OK, I see a few others have been having disk problems with 0.95, but I'll
throw in my experience for what it's worth. 0.95 booted fine, I've installed
it and rebuilt it a few times using the newgcc 1.4. It seems pretty stable,
with the following caveats.
(1) drive problems; both the 'Reset-floppy called' and 'Unexpected HD
interrupt' messages popped up a few times during a several-hour session.
I hardly ever got these under 0.12 (with fast floppy patches). The last
time the floppy problem occurred, I got the continuously scrolling
version, interruptable only by reboot. The last time the HD problem
occurred, it was at the beginning of a make, while cc was fetching cpp; it
obviously made it load an invalid page, since all subsequent invocations
of cpp gave a SEGV. [It would be nice to be able to somehow invalidate all
loaded pages when this problem occurs, to force reloading, although
rebooting obviously solves it.] At least in the case of the HD interrupt,
other things were going on in other VCs (perhaps causing it to miss an
interrupt?)
(2) screen problems; again, something I never got under 0.12 (patched with
virtcons). When scrolling up line by line under uEmacs (`down' key held
down), the screen contents occasionally get out of sync, and also stripes
of strange coloured graphics characters appear at random positions.
Usually, but not always, these problems are transient, and vanish when
scrolling stops. It does this in the standard 80x25 mode as well as others.
But it doesn't do it during standard scrolling, e.g. when doing `ls -lR /'.
setup: pretty standard 386SX/16, 100Mb IDE drive, 3.5 HD floppy, C&T VGA, etc.
Any light on these problems would be much appreciated. Could it be something
to do with using gcc 1.4 to rebuild, instead of gcc 2?
0.95 looks very good, with some useful enhancements; I'm really looking
forward to using the gdb/ptrace combo...
----
Andrew Haylett | Inet: ajh@gec-mrc.co.uk | Fax: +44 245 75244
GEC-Marconi Research | Tel: +44 245 73331 x.3283 | Telex: 995016 GECRES G
