Re: CRC problems

[moon@cambridge.apple.com (David A. Moon)]

> Date: Thu, 16 Jan 92 15:29:04 PST
> From: jbarnett@charming.nrtc.northrop.com
> 
> We have a 3600 that appears to generate a large number of Ethernet packets
> with bad CRCs.  The rest of the packet looks to be well formed.  Since we
> first noted this behavior, the I/O board and paddle have been replaced
> and the power supply checked.  Also, we have tried alternate cabels to the
> Delney box and tried plugging it in to that box at different places.
> Nothing seems to help.
> 
> Questions:
> (1) Where in the software, microcode, hardware, chain is the CRC generated?
> (2) What was the fix when and if you had a similar problem?

I'm pretty sure it's generated by hardwired logic on the I/O board if this
is a 3600/3640/3670/3645/3675, and by a VLSI chip on the I/O board if
this is a 3610/3620/3630/3650.  But maybe there is some way for the cpu to
disrupt the computation of the CRC?  Or maybe it's more likely that the
problem is at the receiving end?