(09-06-2011, 07:28 PM)gid15 Wrote:

Code:

sceKernelRotateThreadReadyQueue priority:0xf is outside of valid range
Error 'java.lang.reflect.InvocationTargetException(java.lang.reflect.InvocationTargetException)' calling hleModule='jpcsp.HLE.modules.ThreadManForUser@600a08', hleModuleClass='class jpcsp.HLE.modules.ThreadManForUser', hleModuleMethodName='sceKernelChangeThreadPriority', hleModuleMethod='public void jpcsp.HLE.modules150.ThreadManForUser.sceKernelChangeThreadPriority(jpcsp.Processor,int,int)'
OnMethod: public void jpcsp.HLE.modules150.ThreadManForUser.sceKernelChangeThreadPriority(jpcsp.Processor,int,int)

This should be fixed in r2301. Could you test again?

Thanks!

(09-07-2011, 07:31 PM)gid15 Wrote:

Code:

sceUtilitySavedataInitStart Address 0x0903D740, mode=1(AUTOSAVE), gameName=ULUS10233, saveName=, fileName=DATA.BIN
hleIoOpen filename = ms0:PSP/SAVEDATA/ULUS10233/DATA.BIN flags = 1 permissions = 00
hleIoRead - unknown id 80010002
sceIoClose - unknown id 80010002

I hope to have fixed this problem in r2302. Does it help here?

(09-07-2011, 11:11 PM)rcoltrane Wrote: I think you are almost there. Here we go (tested with r2302):

Now we have some changes in game behavior because it does not display the memory stick messages anymore, neither the intro video. Instead, it keeps reading the game files (I suppose because the small yellow UMD disc keeps running at the top left corner of the screen) and after some time reading, it crashes. See the .log attached here for the details. This bug is a nasty little one

(09-09-2011, 01:54 PM)gid15 Wrote:

(09-07-2011, 11:11 PM)rcoltrane Wrote: I think you are almost there. Here we go (tested with r2302):

Now we have some changes in game behavior because it does not display the memory stick messages anymore, neither the intro video. Instead, it keeps reading the game files (I suppose because the small yellow UMD disc keeps running at the top left corner of the screen) and after some time reading, it crashes. See the .log attached here for the details. This bug is a nasty little one

Is there something displayed in the command box (java exception)?

(09-09-2011, 03:41 PM)gid15 Wrote:

Code:

Error 'java.lang.reflect.InvocationTargetException(java.lang.UnsupportedOperationException: Converter class com.xuggle.xuggler.video.BgrConverter constructor failed with: java.lang.StackOverflowError)' calling hleModule='jpcsp.HLE.modules.sceMpeg@1b7516d', hleModuleClass='class jpcsp.HLE.modules.sceMpeg', hleModuleMethodName='sceMpegGetAtracAu', hleModuleMethod='public void jpcsp.HLE.modules150.sceMpeg.sceMpegGetAtracAu(jpcsp.Processor)'

There seems to be a stack overflow.
Could you try to set or increase the java -Xss option: e.g.

Code:

-Xss10m

(09-09-2011, 05:11 PM)rcoltrane Wrote:

(09-09-2011, 03:41 PM)gid15 Wrote:

Code:

Error 'java.lang.reflect.InvocationTargetException(java.lang.UnsupportedOperationException: Converter class com.xuggle.xuggler.video.BgrConverter constructor failed with: java.lang.StackOverflowError)' calling hleModule='jpcsp.HLE.modules.sceMpeg@1b7516d', hleModuleClass='class jpcsp.HLE.modules.sceMpeg', hleModuleMethodName='sceMpegGetAtracAu', hleModuleMethod='public void jpcsp.HLE.modules150.sceMpeg.sceMpegGetAtracAu(jpcsp.Processor)'

There seems to be a stack overflow.
Could you try to set or increase the java -Xss option: e.g.

Code:

-Xss10m

I've included this parameter into the "start-windows-x86.bat" file at the commandline string, just like this:

echo Running Jpcsp 32bit...
"%JAVA_HOME%\bin\java" -Xss10m -Xmx1024m -XX:MaxPermSize=128m -XX:ReservedCodeCacheSize=64m -Djava.library.path=lib/windows-x86 -jar bin/jpcsp.jar %*
goto END

But by doing this, the emulator freezes when I push the RUN button and doesn't even start the game. In fact, the emulator screen remains gray, such when you don't select a game from the list to run.
Sorry if I did something wrong, Java is one of my weak points

Anyway, here's another .log attached with the command line above.

(09-10-2011, 11:14 PM)rcoltrane Wrote: Tested again with the latest rev2303. The results were similar: If you don't click any button, it crashes after a couple of seconds by itself. And if you click to access the memorystick, it loads the contents until its complete and crashes right after. Atteched are the commandbox crash report and the new .log file.

(09-11-2011, 09:33 AM)gid15 Wrote:

(09-10-2011, 11:14 PM)rcoltrane Wrote: Tested again with the latest rev2303. The results were similar: If you don't click any button, it crashes after a couple of seconds by itself. And if you click to access the memorystick, it loads the contents until its complete and crashes right after. Atteched are the commandbox crash report and the new .log file.

Thanks for testing new revisions.
The interesting part in the command box is the beginning, which is missing in your report.
I guessed it would be a stack overflow, but I might be wrong. Could you post the beginning of the message in the command box? In the command box menu "properties" you can increase the buffer height if the beginning of the error just disappeared.