Subject: Re: port-amiga bugs, could I help?
To: Petri Koistinen <pkoistin@hit.fi>
From: Ignatios Souvatzis <is@beverly.kleinbus.org>
List: port-amiga
Date: 11/21/2000 20:56:27
Hi Petri,

I think the reason I left the PRs at the stage they are, is noted in
the state change comments in the PRs themselves. In short

On Tue, Nov 21, 2000 at 12:27:16PM +0200, Petri Koistinen wrote:

> For exsample, I could look if PR 2732: "MSDOS fs is broken on the amiga" 
> is still valid.

Right. I suspect it is some spl problem, as writing MSDOS bit format requires
listening to the disk index interupt, but my quick try didn't work.

> And howabout closing suspended PR 8122: "NetBSD not booting" as reporter
> has tried to boot NetBSD with 4 megs of RAM, while the version he tried
> requires 6 megs of RAM.

See there. IMO, it _should_ boot, It just doesn't. The pmap needs to be 
rewritten to be more robust. Thats why it is suspended.

> I guess some feedback should asked from Bernd Ernersti about open PR 2934:
> "kernel panic during a halt or reboot (/dev/reload)". I guess some
> progress may have done been since, and this problem might not any longer
> exists.

Try the problem yourself, first.

> Open PRs 8218 and 8280 are duplicates, could they be somehow merged and
> analyzed?

We have no formal mechanism to merge PRs.

> Maybe amiga portmaster has something new to say about open PR 2649: "A3000
> hangs when attaching HP Scanjet IIcx" which has been submitted by himself.
> =)

If you really want to know: When I ever manage to write bus_dma for Amiga,
I'll try to merge the bus_space/dma sbic driver in some port-arm32 PR, test it
and switch to it. Then I'll try to find out what the problem is.

> PR 5459: "Failures from the rgression tests on an amiga with a 68040 CPU"
> within feedback state has reasonable question at the end: 
> "State-Changed-From-To: open->feedback 
> State-Changed-By: tv 
> State-Changed-When: Sun Nov 5 20:29:17 PST 2000 
> State-Changed-Why:  
> This PR is from ages ago.  Does this still fail in regress?"
> Could I somehow help with this problem with using 68060 or is there
> someone with 68040 able to do the test again. 

Didn't notice. "yes, it faile a couple of weeks ago, no change, no time 
to mentally run the exception handlers in Amiga".

> Is open PR 4332: "PR3123 test program makes 68060 panic if compiled with
> old gcc" still valid, or have gcc been fixed? PR3123 seems to be closed
> and bug described in have been fixed.

Whether gcc has been fixed is irrelevant for the problem. If the cpu allows
it at all, no user program, however broken, should crash the system. 

No, I didn't try recently.

> I think PR 6157: "floppy driver does not always recognize HD disks"  in
> feedback state is fixed. I could test it too.

Note "always". I don't think it is fixed. At least nobody fixed it. You're 
free to investigate, if you have the hardware and the Amiga hardware reference
manual (I lot mine).

> I hope something is done to above probrems and that I didn't make Ignatios
> too angry. =)

only tired.
	-is