Given that we have a problem that can trip up the floppy driver (which is how I
came across the bug in the first place, though I don't know whether to call it
a floppy-driver bug or a kernel bug... that's a whole different argument), my
first response is to just hack the floppy driver to allocate the DMA buffer at
boot time...
Since _this_ hasn't been done yet by ye kernel gurus, perhaps this isn't a
great solution ? Or do people just not like dirty hacks to get around bugs ?
ta for the feedback
Neil Conway
UKAEA Fusion
neil.conway@ukaea.org.uk