Kirk Lewis wrote: > I didn't notice that my reply was too long (due to older posts) and got > rejected from the group. Sorry. Here is my original reply. If anyone can > provide some suggestions on peaking into those memory buffers right > before > and after a DMA write that would be helpful: > > Well that's depressing :*(
Yes it is, you can't imagine the excitement I had at getting to meet the developers and then the depression that set in when I found out they didn't actually have all the great secrets about the chip. It's interesting that internally the real reason for not releasing the documentation and firmware source was because they don't want to reveal the fact of how buggy the chip is (it's well documented on how buggy, without any solutions though), that's why they held it so tightly. It certainly felt better when it seemed the chip had all these secret capabilities and a possible amazing firmware design being held under lock and key, and they knew it by heart every bit of ability and way it worked. That seems to be the big 'white elephant' there, and I am quoting one of the original audio firmware designers on that one :). Thanks, Chris > But there has to be a way to recover. If Anyone > knows how to peek into the memory buffers before a write (the question > I was > asking earlier) that would be helpful. I think I'll also start saving > every > register before a write, then print it out and the change after a failed > write. Maybe that will shed some light on things. After reading > fw-memory.txt it seems not every register is fully understood. Maybe I'll > uncover something. It may be a problem with the hardware, but hopefully > there is a way to work around it with the driver. > > ------------------------------------------------------------------------ > > _______________________________________________ > ivtv-devel mailing list > [email protected] > http://ivtvdriver.org/mailman/listinfo/ivtv-devel _______________________________________________ ivtv-devel mailing list [email protected] http://ivtvdriver.org/mailman/listinfo/ivtv-devel
