Changes were added in lk 2.5.2-pre7 to the advansys SCSI
adapter driver that let it compile. It would seem that
the author didn't have any appropriate hardware to check
the driver (since it dies almost immediately and takes
the machine with it).
This patch http://www.torque.net/sg/p/sg252p7.diff.gz
will hopefully make the advansys driver useful again.
It is not perfect but it:
- runs on a Athlon UP box
- runs on a dual Celeron SMP box
- is running the machine that sent this post
I have been running variations of this patch for the last
3 weeks and it has been sent to Linus and the lkml before.
Perhaps those helpful people who send speculative fixes to
Linus could mark them clearly as "untested".
Doug Gilbert
Em Fri, Jan 04, 2002 at 12:50:38AM -0500, Douglas Gilbert escreveu:
> Changes were added in lk 2.5.2-pre7 to the advansys SCSI adapter driver
> that let it compile. It would seem that the author didn't have any
> appropriate hardware to check the driver (since it dies almost
> immediately and takes the machine with it).
>
> This patch http://www.torque.net/sg/p/sg252p7.diff.gz
> will hopefully make the advansys driver useful again.
> It is not perfect but it:
> - runs on a Athlon UP box
> - runs on a dual Celeron SMP box
> - is running the machine that sent this post
> I have been running variations of this patch for the last
> 3 weeks and it has been sent to Linus and the lkml before.
>
> Perhaps those helpful people who send speculative fixes to Linus could
> mark them clearly as "untested".
That was the case. I said that it didn't had the board. But I planned to
get one that Riel has on his test (iirc) machine to test the patch. Now you
have fixed it, thanks.
I'll take a look at your patch and see if with your changes the
scsi_assign_lock patch that probably will appear in pre8 is still needed
for the advansys driver.
- Arnaldo