ATAPI driver still not working

[Sebastian Sumpf]

Hi Ben,

On 11/03/2014 04:04 PM, Nobody III wrote:
> The ATAPI driver is still unable to detect my hard drive (or my CD drive
> for that matter), even though I have switched to a 32-bit Genode system.
> Any help?

Unfortunately, the ATAPI driver is a really old peace of code. We
haven't used in a long time and there doesn't even exist a proper run
script to test it. I've tried the ones I could find, some seem to work a
little, while others do not work at all. Also there are issues with I/O
APIC using kernels, where one might need the ACPI driver to locate the
corresponding GSI's that replace the legacy IRQ numbers as well as find
out polarity and trigger modes (especially for Fiasco/Fiasco.OC). We
have regarded legacy IDE support as obsolete for a while. I can try, to
at least, build a working run script for some kernel working in Qemu for
a start, but I can't promise, since my time is limited. It seems like
Nova is a good candidate for the time being. So my suggestion would be,
build a run script and find out why the driver does not get an interrupt
(which usually is the cause of all the trouble). The corresponding
driver output should be:

Enabling interrupts:
[init -> atapi_dev Device initialized! Enabling interrupts ...

When receiving the first interrupt:
[init -> atapi_drv] First block: 0 last block <last block>, block size 512

It might be the wrong IRQ number (double check with Linux) or the wrong
trigger/polarity information. If you don't even get to the first output
above, something is really messed up. In that case I would advice you
that you send us the complete output of the test run, your exacts
hardware specifications, and if possible a working run script.

Regards,

Sebastian



-- 
Sebastian Sumpf
Genode Labs

http://www.genode-labs.com · http://genode.org

Genode Labs GmbH · Amtsgericht Dresden · HRB 28424 · Sitz Dresden
Geschäftsführer: Dr.-Ing. Norman Feske, Christian Helmuth