From: Frank C. E. <fe...@ai...> - 2001-07-27 05:35:14
|
> There's one strong argument against this theory: > Assume that there are disturbing GUI-commands wich are send in the time > between kick-off and idle check that are supposed to fire of DMA again. > I comment out the kick-off line in the code. By our assumption the > DMA-transfer must be kicked-off though. But it isn't. The engine stays > idle. Contradiction. Therefore this hypothesis is wrong. Or did I miss > something? The symptoms are the same, but could be caused by differing things. While we did have that problem in Utah, we _shouldn't_ have it in the driver we're working on- I suggested checking for the bus mastering being disabled to eliminate the possibility. The disturbing thing with what we've got right now is that register always being zero on my machine with the code- it should accept the information implicitly. It's almost as if we're writing to something other than the MMIO area- at least with my system setup. I'm going to look into how the code gets that mapping we're using and how it's supposed to be used. Gareth, any insights here would be welcome... :-) -- Frank Earl ------------------------------------------------------- |