Ximenes Rocha Resende (ximenes@cv-2-8.campusview.indiana.edu)
Fri, 29 May 1998 17:11:44 +0000 (UTC)
Here I am again... Well, still trying to have SparQ working in my system.
I can load paride,epat(port 0x378 mode 5) and pd and have them read the
content of my disk. The problem is when it comes to writing it. After a
while the kernel flushes the message
pda:do_pd_write_drq:state = 0x10050 = SEEK READY TMO
continouslly. Trying to "debug" the drivers' source code I figured out that
the problem appears to happen within the "while" loop at "pd_wait_for"
the driver's code...
If somebody has SparQ running ok with paride 0.94, please send me a message
saying so. I need to be sure that the driver's has no bug. I'm beginning to
suspect that my hardware has problems (despite the fact that Sparq didn't
show any problem under WinNT yet!). Thanks, Ximenes.
PS: My system is not hanging as I thought before. It' just extremely slow
because it's going through the loop in "pd_wait_for", processing sector by
sector slowly...
XIMENES ROCHA RESENDE
Department of Physics Swain Hall West 117
Indiana University Bloomington IN 47405
Phone: (812) 855-3152 Fax: (812) 855-5533
http://cv-2-8.campusview.indiana.edu/~ximenes
On Sat, 16 May 1998, Grant R. Guenther wrote:
> > No, I commented out the line in the linux source (in Makefile) that set SMP.
> > Then I compiled 2.1.98.
> > By the way, I changing my CMOS settings and now I am using EPP. But the situa
> > tion didn't improve. I was thinking about what you said, that maybe my sys
> > tem's memory was low... I did the following experience:
> >
> > - Booted my system and run "top q" under X.
> > - Mounted my SparQ and copied a huge directory from my HD to it: /usr/X11R6
> > (app. 23 Mb). As the files were copied I kept watching how much free memory
> > the system had. A very abrupt jump happened: from 44M to 20M, at which value
> > the system hung.
>
> Are you sure it actually hung ? When that happens, the system goes off to
> try to flush the buffers to disk - which will take a while - 60 seconds
> or so in EPP mode with the default "delay" setting.
>
> > This happens for both linux-2.0.32 (Paride 0.94) and linux-2.1.98. Do you think
> > it might be related to memory leak in the kernel (block_write -> getblk) ?
>
> No.
>
> If you intend to use a PARIDE drive and X at the same time, it might be a
> good idea to use the "nice" option when you load pd.
>
> --------------------------------------------------------------------------
> Grant R. Guenther grant@torque.net
> --------------------------------------------------------------------------
>
-- To unsubscribe, send mail to: linux-parport-request@torque.net --
-- with the single word "unsubscribe" in the body of the message. --
This archive was generated by hypermail 2.0b3 on Wed 30 Dec 1998 - 10:17:47 EST