Lunar Eternal Blue (JP) doesn't work

[GoogleCodeExporter]

What steps will reproduce the problem?
1. Load any image of Lunar: Eternal Blue (here is the cuesheet used in this 
case - http://pastie.org/6002365)
2. When I start up the '.cue' file, it shows me the Japanese Mega CD screen. 
After pressing Start, it shows the audio CD screen.
3. I then press Start again with the cursor highlighting the 'CD-ROM' button. 
It shows a 'Sega copyright screen'. After that, nothing will show up and the 
game won't progress - screen stays black.

What is the expected output? What do you see instead?
Expected output should be the game running. What we see instead is a black 
screen after the Sega copyright screen.

What version of the emulator are you using (official, SVN revision,...)?
Latest SVN revision.

Please provide any additional information below (Emulator settings, Console
setting,...)

I'm trying to load the Mega CD game 'Lunar - Eternal Blue' with a CUE sheet 
that looks like this -

I'm curious what the issue here is. I've had another guy test a bunch of 
Japanese Mega CD games for me and so far everything works except for Lunar 
Eternal Blue. Is there something wrong with my libretro code in libretro.c?

According to a user on my forum -

http://forum.themaister.net/viewtopic.php?pid=1360#p1360 (it's the first issue 
he mentions in this post but it seems he wrongly misattributed it to the 
libretro port).

According to him, the Japanese version of Lunar: Eternal Blue used to work, but 
not anymore. He initially thought that it was libretro's fault - but it turns 
out it affects the Wii version of Genesis Plus GX too.

Original issue reported on code.google.com by libre...@gmail.com on 2 Feb 2013 at 2:02

[GoogleCodeExporter]

it's already fixed in r758, it was a regression introduced in r757, which was 
done after the official 1.7.3 release, hence why your user said it worked fine 
in the stand-alone version

As a technical reminder, when i fixed read-only registers on MAIN-CPU side in 
r757, I forgot about PRG-RAM write-protection register, which was then never 
set anymore. It appears this game accidentally tries to access the 
write-protected area, which ended up corrupting the SUB-CPU side code in RAM 
and later cause an exception.

Original comment by ekeeke31@gmail.com on 2 Feb 2013 at 9:11

• Changed state: Verified

[GoogleCodeExporter]

Thanks ekeeke.

BTW - as you can tell from the countless mails you've probably received from 
that nintygaming guy - people are not satisfied with how RetroArch GX currently 
looks like, and me and ToadKing are seriously not sure whether we are even 
doing things right in that regard anymore.

Could you go over the RetroArch GX port and possibly look at gx_video.c and 
tell us if/what we are doing wrong? I also heard from a guy that setting the 
resolutions the way we do on GX caused adverse performance in HBC and even the 
Wii Menu - so this has left me feeling very worried about the entire thing.

I would feel much more comfortable if somebody experienced like you could take 
a good look at it and help me out there in that regard.

Original comment by libre...@gmail.com on 3 Feb 2013 at 8:54