Hi Devin, > Do you have the USB trace you used as the basis for this dump? And > did you programatically parse that dump to generate the table, or did > you enter it by hand? Could you please email me the underlying dump > so I can double check the data? I wrote a program to parse the trace and make it more readable, but I entered the table by hand. Unfortunately, I no longer have any qam traces. I can make a new one, if you'd like, but it'll have to wait until I have time to install windows. > I think the correct approach here would be to have three tables, a > table of common registers, and two tables, one for each modulation > type. It would program the common registers first, and then pick the > correct table to program the rest. If we are concerned about the > order that the registers are being programmed in, we can just setup > the two tables so they start at the first register which is different > (since it is almost at the bottom of the table anyway). I didn't notice how similar they were when writing the patch, thanks for pointing it out! I'll do some tests and see how much can be combined into a common table. While I'm at it, I thought I'd go ahead and make a patch to remove the top bits from the vsb table, but I've got a question about that. I think the first four entries are unnecessary. I'm pretty sure 8090 and 8091 have to do with the 8522's i2c controller and 4092 is a status register. I have no idea what 2005 is, but the new code would change it to A005 and I don't remember seeing either in any of the traces I did (though I never did a vsb trace). Is this correct or am I missing something? If I make a patch, can you or someone else test it for me? (can't get a signal here) Frank -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
