I thought I would just start a new thread about this card and my experiences with it rather than clogging up any of the existing threads about the micro and Sirion USB.
I recently purchased a multi function PCI card for the microA1 that has 4 usb2 ports, based
on the VIA VT3104 USB2 controller and a VIA VT6421 IDE/SATA controller on one card. You can
view it here: http://www.sybausa.com/productInfo.php?iid=801
Note - I found that unless I tighten down the card with the slot connector screw, I would get
lots of lockups when I plugged in or unplugged USB devices.
To get this card to "work" with OS4.1 update 5, I do the following:
In startup-sequence I comment out the following:
;If EXISTS S:Network-Startup
; Run >NIL: <NIL: *>NIL: Execute S:Network-Startup
This prevents the 3com ethernet device from being loaded. As pointed out by someone else
in this group, there is a contention between the 3com therent device and the EHCI module
of Sirion....at least, that's how it looks. I guess it could also be a hardware/interrupt level issue
perhaps on the microa1/AmigaOneXE.
I turn off the intenal USB0 and USB1 ports of the micro A1 in uboot.
I then powerup the micro A1.
Yes, all the USB kickstart modules are active, even EHCI.
If I insert any USB1.1 device into any of the four ports of the multicard, I see
the popup showing it has been recognized and installed. It then becomes
accessible - printers, usb mice, even USB to serial converters are recognized
(though the usb to serial converter does not have a driver).
If I plug a 2Gig flash drive into the multicard, I get a few different possible
results as follows (note tht this flash drive works fine on the internal USB1.1 of the micro
under normal operation):
#1 Sometimes, the flash drive is recognized after about 3 to 10 seconds, during
which time the USB attachment window shows blanks for vendor, product and class,
and "locating device" underneath. Then the device is found. It is mounted and can be
used, but is VERY slow, far slower than USB1 speed. It often seems to hang up for
Here is a usb.log from when that happens.. I have plugged in a USB mouse and a USB
printer (or serial converter) and then the flash drive is plugged in to port 3:
#2 Sometimes the flash drive shows the USB attachment window with all blank entries,
Hidden Text - Click to Show :
and "locating device" underneath, and after 30 seconds closes. The device is not found.
Here is the usb.log of when that happens:
#3 Sometimes I see the flash device on the workbench, after about a minute, showing
Hidden Text - Click to Show :
usb0:NDOS - (or similar). the device is mounted but not recognized.
If I plug in an Aluratek USB DVD/rw device, which uses 2 USB ports, I get similar
results as with the USB flash card. Sometimes it is seen properly and works VERY
slowly, sometimes it is not seen at all.
Usbctrl RESTART does not help. If I do a usbctrl restart, nothing happens.
If I unplug the USB flash or USB DVD, usbctrl restart still does not fix any issues.
Whether or not any USB devices are plugged into the multicard, when I execute the
network-startup and invoke the eth3com.device, the Amigaone locks up. The
pointer no longer moves, but CTRL-AMIGA-AMIGA still works. Programs running such
as AmigaAMP lock up and the music just stays at whatever tone was playing.
Once in a great while I can run network-startup and it works just fine as long as no USB
devices are plugged into the multicard - not sure why it works one in 10 times or so.
I have tried adding to uboot the environment variable usb_ohci_orphan_checking 1
and this does not seem to help in any way. Is this an implemented function for the microA1?
The USB controller chip, the VIA VT6214L is the 4 port version of the VIA VT6212L that others
on this forum report has been found to work in some cards (in the 2 port version VT6212L).
If I comment out the kickstartlayout line:
and reboot, then I can run the network AND the USB (as 1.1 of course) without any lockups.
If there is any way I can help debug this problem...please let me know! I would love to get this cards
USB2 ports working...and it is soooooo close! If anyone wnats me to test a new ehci kernel module with
this card, please let me know. I'm willing to sign an NDA or whatever to help out.
I have many more log files I can provide....