On Tue, Dec 06, 2005 at 03:33:23PM +0100, Michael Schmitz wrote: > On Mon, 5 Dec 2005, Johannes Berg wrote: > > Following the hints at > https://lists.berlios.de/pipermail/bcm43xx-dev/2005-October/000305.html, > I extracted the necessary offsets from the driver file. The iv offset > I simply guessed to be 0x18f0 before the first ucode chunk. Is that safe > to assume? Some of the resulting iv .fw files were empty ... how can the > offset of the iv be determined?
Please, don't think that it's based on a fixed length. That's utterly wrong. After looking into a bunch of driver binaries I have a sense for it now. So, if anyone is unsure about some offsets please drop a note... Your file with md5 86cc708e8df3b035a1dbea41ac4eb7d2 is now supported by fwcutter. > A question for the always cautious Ben: can wrong iv values damage the > card? Definitely, you shouldn't try wrong iv (or other) offsets!! Nobody from us knows the details from this chip. We only know how the chip works with some mips and macosx drivers. But that'll never be enough to gave warranties in case of such a random data flood. Don't do it, please! martin _______________________________________________ Bcm43xx-dev mailing list [email protected] http://lists.berlios.de/mailman/listinfo/bcm43xx-dev
