SCSI lockups !?

Thread View: comp.os.linux.hardware

1 messages

[T] Back to Threads [V] Tree View [G] Group Articles

1 total messages Started by Graham Kean Tue, 10 Dec 1996 00:00

#376

Author: Graham Kean
Date: Tue, 10 Dec 1996 00:00

40 lines
2234 bytes

Can anyone suggest why linux would lock up when copying large
amounts of data to a PCI/SCSI disk ?  I saw a reference to this
regarding 1.3.x kernels, but I have 2.0.26.

I have a DPT PM2044UW (Smart Cache IV, Ultra-Wide), connected to a 
single IBM 2 Gb drive.  When copying (cp -dpR) from one partition 
to another, or across NFS, the system would lock-up after about 
10-20 Mb had copied.  One one occasion I got continuous console 
errors like "eata-reset: slot 1 locked, did reset ...".

A power-off or hard reset was required which of course damaged the 
filesystem.

I seem to have cured the problem by configuring the DPT BIOS so
that all SCSI transfers are asynchronous, and there is no command
queueing (max simultaneous cmds = 1).  Might this mean:

    1. The Linux SCSI driver doesn't support syncronous mode or
       command queuing ?

    2. The kernel should support it but has a bug ?

    3. The motherboard doesn't support DMA properly ?

    4. I have a hardware fault ?

    5. Something else entirely ?

And what have I done to the performance of the disk system by 
turning off the useful features ?

Any ideas gratefully accepted.

-- 
--------------------------------------------------------------------
 Graham Kean                                  g.kean@auckland.ac.nz
 University of Auckland Medical School
 Private Bag 92019, Auckland, New Zealand.
--------------------------------------------------------------------

Thread Navigation

This is a paginated view of messages in the thread with full content displayed inline.

Messages are displayed in chronological order, with the original post highlighted in green.

Use pagination controls to navigate through all messages in large threads.

Back to All Threads