On Mon, Oct 18, 2004 at 10:16:19PM +0300, Jukka Marin wrote:
> Sometimes it makes me wonder what the purpose of NetBSD releases _is_.

Well, I don't know enough to say whether this is currently the case,
but I know in the past, NetBSD has had a "do it the right way"
attitude which has occasionally caused it to not apply quick hacks for
known issues, but instead wait for a real fix to be developed. One
example that comes to mind is ISA DMA bounce buffer support. A lot of
people waited a long time for an official release that let them use
their Adaptec 1542 SCSI cards in machines with more than 16 megs of
RAM... if I recall correctly, by the time NetBSD supported it
(mid-1997), the 1542 was already on its way to obsolescence. However,
unofficial patches had been available for years for those who wanted
them--they wouldn't help for those who wanted to use a 1542 in their
DEC Alpha, unlike the official fix, but the majority of users had
i386s.

The umass issue seems very similar--a patch is available for those who
want them, but I'm expecting a true fix will come that targets the
root cause of the problem (i.e., one that will also solve the issue
some are having plugging in Cardbus cards, in addition to the umass
flash disk problem).

So in any case, while it would've been nice to get a real fix in, I
don't see the current situation as a problem.
-- 
Name: Dave Huang         |  Mammal, mammal / their names are called /
INet: khym@azeotrope.org |  they raise a paw / the bat, the cat /
FurryMUCK: Dahan         |  dolphin and dog / koala bear and hog -- TMBG
Dahan: Hani G Y+C 28 Y++ L+++ W- C++ T++ A+ E+ S++ V++ F- Q+++ P+ B+ PA+ PL++