Michael Ditum wrote: > Whilst we've fixed the recurring IRQ issue some times we still get the > NMI Issue that I first talked about. Sometimes when I boot up I get > this problem other times I do not. I've run memtest86+ for one pass > and it came up with nothing... I'll try a longer run during my lunch > break just to make sure but I don't believe it is a RAM issue. NMIs are usually caused by broken hardware. > According to the stacktrace the last time it was in the dvb code was > in start_ts_capture in budget_core.c. I believe its these lines... > > 113: saa7146_write(dev, MC2, (MASK_08 | MASK_24)); > 114: mdelay(10); > 115: > 116: saa7146_write(dev, BASE_ODD3, 0); > 117: saa7146_write(dev, BASE_EVEN3, 0); > 118: saa7146_write(dev, PROT_ADDR3, TS_WIDTH * TS_HEIGHT); > 119: saa7146_write(dev, BASE_PAGE3, budget->pt.dma | ME1 | 0x90); > > Why is there a 10 millisecond delay done there? Don't know. Imho it is not required but it doesn't hurt. > How could it cause an NMI? It can't. :D > Is it OK to comment out that line? Imho yes. > I appreciate any help in tracking this issue down. Try your DVB cards in a different machine. Oliver -- -------------------------------------------------------- VDR Remote Plugin available at http://www.escape-edv.de/endriss/vdr/ --------------------------------------------------------
