I know, I probably gave a lot of people heart attacks just from that title. I'm not saying that I would actually try flashing my BIOS with WINE, I was just wondering, purely out of theoretical curiosity, if it's even possible to flash the BIOS by initiating an "exe" BIOS flasher from within WINE. I would suspect not, given that WINE "can't even" run Windows drivers, and from my novice viewpoint, to do this would seem to be even lower than a driver. That said, you never know. However, WINE does manage memory, I've heard. Doesn't an exe BIOS flasher (when initiated from Windows) just write some stuff to RAM and send some reboot instruction to the processor or something? If WINE can manage a program's memory, couldn't it write the BIOS flasher stuff to memory, and then tell the operating system to send that reboot command to hardware? Or are all WINE processes' memory housed within WINE's allocated memory, and thus would take special WINE development to be able to add BIOS flashing support? If the dev team wanted to work on it, I would imagine there'd be a safe way to try it out. Perhaps you can take a BIOS image file, copy it, and tell QEMU or something to use the copy in "read and write" mode, install Linux in QEMU (or whatever), put in your magic developing stuff, and try to flash the BIOS. Of course you want to make sure you're running on an emulated processor, not your real one. QEMU does have a -bios option to "set the filename for the BIOS", and it looks like that's how the folks at openfirmware are trying things out, though I don't know if QEMU supports _writing_ back to the file. Cheers, Jake
