Bug-report: reading colorram in turbo-mode [1 Attachment]

[Mikkel Hastrup enkorv-/[e-mail] [chameleon_64]]

Hi guys,

I've found a bug, not sure if it has been reported or detected already (but doesn't seem to be mentioned in the mailinglist atleast).

I've noticed - when playing around with the turbo-mode with latest firmware - that reading from colorram seems to be broken.

Setup: Chameleon (Beta-9a, rev 753) cartridge mode on a C64C PAL.
TLR's Box Check Type (with 1541u2 cartridge) says: C64 DL 6569 6581 6526A 6526A.

This bug appears atleast on the VIC-output, I haven't checked on the VGA-output.

I've included a test-program (+ source) with some colorram copying. First section on the screen is written and copied in '2 mhz mode' (so Turbo $d030 bit needs to be enabled). The other section is in '1 mhz mode'. It should look identical in both sections but on my setup, only the second looks correct. Anyway, the code is quite self-explanatory.

[enkorv-/[e-mail] [chameleon_64]]

...and since the attachment got corrupted:http://undone.se/misc/test_chameleon.zip http://undone.se/misc/test_chameleon.zip

[Peter Wendrich [e-mail] [chameleon_64]]

Hoi Mikkel,

Yeah, Tobias has mentioned the same issue to me last week when he was
optimizing the menu code. I hasn't been reported on this list. Anyway
I've a fix ready for this bug and is planned to be included in the 9b
release. I've also added a test-case to our release test suite, so this
bug will hopefully not happen again.

Thanx for reporting!

L8r,
Peter

[Attachment(s) <#TopText> from Mikkel Hastrup included below]
Hi guys,
I've found a bug, not sure if it has been reported or detected already
(but doesn't seem to be mentioned in the mailinglist atleast).
I've noticed - when playing around with the turbo-mode with latest
firmware - that reading from colorram seems to be broken.
Setup: Chameleon (Beta-9a, rev 753) cartridge mode on a C64C PAL.
TLR's Box Check Type (with 1541u2 cartridge) says: C64 DL 6569 6581 6526A 6526A.
This bug appears atleast on the VIC-output, I haven't checked on the VGA-output.
I've included a test-program (+ source) with some colorram copying.
First section on the screen is written and copied in '2 mhz mode' (so
Turbo $d030 bit needs to be enabled). The other section is in '1 mhz
mode'. It should look identical in both sections but on my setup, only
the second looks correct. Anyway, the code is quite self-explanatory.

------------------------------------
Posted by: Peter Wendrich <pwsoft-***@public.gmane.org>
------------------------------------