https://bugzilla.redhat.com/show_bug.cgi?id=1893631
--- Comment #8 from Marc <marc.payne(a)mdpsys.co.uk> ---
Todd,
Ahhh, the CH382L. I have a card based on that chip here. This news has left me
surprised, confused and disappointed – all at the same time!
I was expecting the chip to be the MosChip MCS9901. I could see the MosChip
logo on the photo of the card on the Lava website and they appear to sell other
cards using MosChip chips. From the lspci output you’ll notice that you have a
Serial controller in addition to the Parallel controller. The MCS9901 is a
lower cost, more limited version of the MCS9900 (the chip on your SIIG card).
One of the limitations is the modes of operation – you can’t have the parallel
port enabled without having at least one serial port enabled. Since there is
no serial port physically present on the card, Lava have set the device ID of
the Serial controller to 0xDEAD. The serial driver won’t recognise the device
so no ghost /dev/ttyS… device will show up on the system. The subsystem IDs
A000:1000 and A000:2000 also match the defaults used by MosChip.
Did you buy the card recently? Was it direct from Lava? I assume it looks
physically different to the one pictured on the Lava site? Perhaps they have
changed the chip they are using… I can’t explain that lspci output from a
CH382L-based card.
My (admittedly brief) experience of the CH382L is not great. Both the
datasheet and the manufacturer’s website indicate that the parallel port should
support the faster data transfer modes defined by IEEE1284 – EPP and ECP. My
card only supports the legacy software controlled SPP mode and that makes it of
limited use. It briefly crossed my mind that I might have a defective or fake
chip but after downloading and examining the Windows driver direct from WCH I
can see that my card is working as it should. Not sure what to say about that…
Marc
--
You are receiving this mail because:
You are on the CC list for the bug.