Hippoplayer GURU on module playback

[alcomatt]

My beloved Hippoplayer have recently started crashing on my fs-uae emulated environment. (A1200 with Blizz IV 30/50, 64MB Z3 RAM, 2G Chip, Kick 3.1.4, PFSaio partitioned physical ssd, run via rdb mode in the emu)

The app will reset Amiga (GURU) on any module playback. The same modules play in other apps. GURU no. is non specific, just relating to memory allocation as it often is.

80000003 1084B600

I SnoopDossed as much as I can but I cannot pinpoint any suspect library or file.

It was working absolutely fine in the past and I am at a loss why it has gone wrong.

I have tried various emulated Amiga configs (down to barebone A1200 with 8MB Fast) and yet no change in the behavior.

Unfortunately I cannot test on real amiga atm as it is being recapped...

Perhaps someone have came across this issue before. I am beginning to pull my (very few) remaining hair out. I'd rather not have to re-build this hd install from scratch.

Thanks in advance.

[Samurai_Crow]

Have you tried re-downloading Hippoplayer from Aminet.net to verify it's not a corrupt executable?

[alcomatt]

Originally Posted by Samurai_Crow

Have you tried re-downloading Hippoplayer from Aminet.net to verify it's not a corrupt executable?

Thanks for the reply!

Indeed i have tried that. No luck. Also replaced Hippoplayer groups files etc basically everything from the lha archive of aminet. The actual app starts up ok. I can see the gui, pick up any module and it instantly crashes. The same module will open happily in protracker and play without an issue.
Tried both OS and AHI replayers, same difference.

[alcomatt]

I have managed to solve this mystery. It was the hip.key file from aminet in L: that was the culprit. Upon removing it from its location Hippo started to work again... happy days!

[andy2004]

have you looked in to the hippoplayer.key file in the L: folder.. use the one from the aminet
NOT from the hipkey generators. that will cause hippo to crash.. tested on my emulation.

[alcomatt]

Mine must have gotten corrupted because all started working once I had removed it. Bizarre...

Thanks for all the replies and suggestions!

[koobo]

I just recently went through the (mess of a) source code and found that the past me had implemented, among other clever things, some keyfile validity checks:

    
        tst.l	keyfile+40(a5)	
	beq.b	.zz
	move.l	tempexec(a5),a1
	addq.l	#1,IVVERTB+IV_DATA(a1)
.zz

This basically just corrupts the exec interrupt structures causing an instant crash, if the keyfile is not valid according to this test. There are several others, too. Sorry