[PARPORT] System ressource pig??


Gunter Magin (magin@skil.camelot.de)
Fri, 6 Mar 1998 11:51:19 +0100 (MET)


Hi,

I have set up Thomas Heptner's (he posted some weeks ago on this list) IDE
CDROM with a Kingsbyte controller on the parallel port of my computer. I
used the paride-0.94 suite, as recommended by Grant. My motherboard is a P90
Premiere-II (Plato) with a SMC37C665 IO chip, which is supposed to be ECP
and EPP compatible.

System logs are:
Mar 5 22:20:52 skil kernel: paride: version 0.94 installed
Mar 5 22:20:57 skil kernel: paride: k951 registered as protocol 0
Mar 5 22:20:57 skil kernel: paride: k971 registered as protocol 1
Mar 5 22:21:06 skil kernel: pcd0: kbic 0.94, KingByte KBIC-951A at 0x3bc,
       using mode 2 (8-bit)
Mar 5 22:21:09 skil kernel: pcd0: Slave: 86803I SE

To check stability of the setup, I read the total contents of a CD on hard
disk to verify it against what a cdrom on the local IDE bus has read. That
test was successful, so no data has been altered or dropped. (I had a
trouble with a bad cable the other day). The achieved data rate was about
100 - 130 KByte/sec. The HD was a SCSI type with busmaster DMA, so it
should have minimal impact on the performance.

During the read from the pcd device I have experienced heavy system
performance degradation: most of the cpu cycles seem to have been burnt in
kernel mode (>80% system load). The system's responsiveness on user input,
both mouse and keyboard, was really bad: 10s of seconds without reaction.
It seemes to me the parport driver suite excessively disables interrupts,
and probably polls a lot around, without yielding to other system
participants. Even the clock interrupts must have had quite some overruns,
as I have noticed after about 2 hours heavy PP-CDROM usage the (SW) clock
(-> date) was 2 minutes behind the (CMOS) clock (-> clock) ! Both clocks
are in synced at system boot time.

When mounting the pcd device, I noticed some error messages:

Mar 5 22:21:27 skil kernel: pcd0: read block completion: alt=0x51 stat=0x51
        err=0x20 loop=0 phase=3
Mar 5 22:21:27 skil kernel: pcd0: read block data done: alt=0x51 stat=0x51
        err=0x20 loop=0 phase=3
Mar 5 22:21:27 skil kernel: pcd0: read block completion: alt=0x51 stat=0x51
        err=0x68 loop=0 phase=3
Mar 5 22:21:27 skil kernel: pcd0: read block data done: alt=0x51 stat=0x51
        err=0x68 loop=0 phase=3

During reading, there are no error logs.

Do those err=0x20 err=0x68 have anything to do with the resource
utilization and the responsiveness?

Is the achieved data rate normal, or is there a way for higher
throughput?

Thanks
gm

-- 
Gunter Magin                                        magin[AT]skil.camelot.de

-- 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 - 10:17:31 EST