The various varieties of ARM architecture hardware most certainly don't have
a standard set of mappings from common module aliases `soundcore', etc.
onto modules; for this reason, it seems we need a system like m68k, where
we'd have /etc/modutils/arch/arm.{netwinder,acorn,...}.
There's a bash function archmodel() in update-modules:
archmodel() {
local arch=`arch`
local model=""
if [ $arch = "m68k" ]; then
if [ -f /proc/hardware ]; then
model=`cat /proc/hardware | sed -ne 's/^Model:[[:space:]]*//p'`
case $model in
Atari*) model="atari"; ;;
Amiga*) model="amiga"; ;;
Macintosh*) model="mac"; ;;
Motorola*) model="MVME"; ;;
*) model="generic"; ;;
esac
model=".${model}"
else
echo "/proc/hardware does not exist, assuming general m68k system"
fi
fi
echo "${arch}${model}"
}
I assume the same wants to be done with the `Hardware:' field in /proc/cpuinfo
on ARM architectures -- the question is, what should the canonical mappings
be?
If both the A5000, RiscPC and later hardware can all be accommodation under
one set of modules, I would propose making the sub-architecture simply
`acorn', denoting all hardware derived from 32-bit Acorn ARM machines.
However, this possibly may not be the case; how about `a5k', `medusa', and
then other aliases files for whatever weird and wonderful hardware Castle
are dreaming up (`mico', `riscstation')?
What about the Footbridge platforms?
c.
