erik quanstrom wrote:
But it IS a bit frustrating to see drivers available in one F/OSS OS (or
variant) and not another, more especially as they are nearly always written in
reasonably portable 'C' code these many years.
that's easier said than done. Blocks are not the same as sk_bufs.
for that matter, they're not the same as MsgBufs or RingBufs.
Aware.
But take note of Haiku's move to make their OS capable of using *BSD drivers.
That sort of adapter layer seems worthwhile, even if the results do not
initially approach 'native' performance.
*After* the dust settles (who woulda thunk, on apparent 'merit' or lack thereof,
that the dodgy Realtek silicon would ever have become soooo ubiquitous?)
..THEN 'native' drivers could be gone after for the much smaller subset of
'common survivors'. IOW, better a slow, or feature-stripped driver than none at
all.
Reality is that the rate of introduction/change of hardware/silicon is too fast
for any small - or even 'medium sized' team (FreeBSD for example) to keep up
with on their own... and that gap is widening.
that's only a problem if one's goal is to support all possible
hardware.
- erik
Not a goal I'd seek - nor even lcose to it.
It is not really a problem if, for example, one says there is only a VESA 2 (or
later) driver for video - so long as it is a good one, as Plan9 is not really a
GUI-centric OS in the first place. More akin to ncurses with a precocious mouse.
;-)
Bill