In article <199805200122.SAA18586@Kowhai.Stanford.EDU>
	jonathan@DSG.Stanford.EDU wrote

>> I have to recheck why my case had
>> hideous throughput. (busy network, sluggissh NFS server)
>
> I'm not aware of the ioctl asic serializing
> scsi and lance access at more than the 16-byte chunks in which the
> LANCE sends packet data out its pins, or possibly individual DMA
> transactions from the SCSI-bus, so I dont think that's a factor.

I see.  In the conext of MI SCSI improvement I will have to run close
analysis what's happening around IOCTL asic activity at heavy loads.

The NFS server was 175MHz Alpha AXP DEC3000/600 with 256MB memory and
RZ26L disk.  OS was NetBSD/alpha 1.3 release.  I feel it runs faster
than Digital UNIX, but wondering it can run even faster.  Ahh, someday
I should pay attention for Prestoserve integration to NetBSD kernel...

>> By the way, I'm occationally experiencing process lockup symptoms with
>> parallel kernel make. 
>
>  ... snipped ...
> from compiles on a -current kernel dating
> back to March, I would ahve said "never".

I've experienced the troubles since the days backto 1.2D or some.  My
kernel is based on 1.3 release (with local hack).  My current theory
on the cause is VFS locking on directories.  I could not succeed in
finding the simpliest condition I can reproduce them.

> Is this also on NFS?

No idea, because I've seldom ran parallel kernel make over NFS.  I
will check it too.

Tohru Nishimura
Nara Institute of Science and Technology