FERGUSON@TMASL.EXXON.COM (04/07/89)
I'm barely (if at all) familiar with GPIO calls, and I'm having a significant difficulty with them and my National Instruments IEEE-488 Bus card. It seems that something causes the call pbu_$enable_device to stop, and spin its wheels for an indefinite period. The only way out is CNTRL-Q. There is no possibility of setting a timeout for this call, and I'm at a loss. The support people at National Instruments have no idea what's going on, and don't seem to care (we already paid for the device, why bother helping us?) Does anyone have a similar experience with this board, or similar experience with the GPIO calls? It's a DN4000, sr9.7, National Instruments IEEE-488 instrument bus card and software. Oh, one more thing. High disk and CPU traffic on the node almost guarantee the problem to happen. Thanks to anyone who can help. Scott Ferguson ferguson@erevax.bitnet (201) 730-2339
dbfunk@ICAEN.UIOWA.EDU (David B. Funk) (04/08/89)
WRT posting 8904071351.AA02833@umix.cc.umich.edu>,
Scott,
You say that you are having GPIO problems with a DN4000 & sr9.7,
two questions:
1) Is it sr9.7 or sr9.7.some.thing (such as sr9.7.0.4 or sr9.7.1.2)
There is a patch tape out to fix some bugs in sr9.7 (no suffix)
related to I/O & DMA problems. If you have sr9.7 (no suffix), I
recommend sr9.7.0.4 (patch #133). Don't use sr9.7.0.6 (patch #143),
it introduces a bug that can cause it to crash when you try to
mount a floppy volume.
2) Do you have the correct GPIO manual? You should be using manual
000959 Rev 10, plus the DN4000 update pages (or newer?). The older
manuals (including rev 10, sans update) only have the DN3000 GPIO
calls. The DN4000 (and DN3500, DN4500) must use a different set
of GPIO system calls for DMA because it has an I/O page map.
Dave Funk