> The problem was addressed with your help, and I do really appreciate your > help.
Great! I hope you learned your lesson though: when something breaks, make sure to re-test using the head of the git repo. Of course that may break something too, but most other developers will also be working from the head, so can probably reproduce your problem. (Or will be more likely to listen to any complaints). Also its worth writing a couple of simple drivers; one that maps a PCI IO memory space and another that maps PCI memory space. That way you can cause the machine check in a very basic driver, i.e., your problem had nothing to do with your specific card. Take a look at the LNX-723 document and code, I wrote it while using a Yosemite board, but have moved on to a Freescale MPC8349E device, since the 440EP wasn't going to work for my application: http://www.ovro.caltech.edu/~dwh/correlator/cobra_docs.html Here's a copy of my test notes: http://www.ovro.caltech.edu/~dwh/powerpc_440ep.pdf Cheers Dave _______________________________________________ Linuxppc-embedded mailing list [email protected] https://ozlabs.org/mailman/listinfo/linuxppc-embedded
