WHDLoad MantisBT - Blastar
View Issue Details
0003779Blastar[WHDLoad Installs Games] slavepublic2017-12-21 12:502020-04-08 00:04
ReporterWepl 
Assigned ToJOTD 
PrioritynormalSeverityminorReproducibilityalways
StatusassignedResolutionopen 
PlatformOSOS Version
Product Version 
Target VersionFixed in Version 
MachineA4000
CPU68060
CPUSpeed60
ChipSetAGA
GFXCardCyberVision 64/3D
ChipMem2 MB
FastMem128 MB
WorkbenchOS 3.9
KickROM40 - Kick 3.1
KickSoftSelfmade
WHDLoad18.4beta
imported
Summary0003779: interrupt acknowledge problems, requires NoCache on 68060 to run
Descriptionwithout NoCache: in the first level screen is trembling, maybe caused by interrupt problem
dump is with ChkInts and fault is occuring before game screen appears
works fine with NoCache
TagsNo tags attached.
Attached Files? blastar.whdl_register (3,740) 2017-12-21 12:50
http://mantis.whdload.de/file_download.php?file_id=588&type=bug
? Blastar.slave (1,508) 2017-12-21 14:15
http://mantis.whdload.de/file_download.php?file_id=589&type=bug
? blaster.2.whdl_register (3,678) 2017-12-21 14:58
http://mantis.whdload.de/file_download.php?file_id=590&type=bug
? Blastar-2.slave (1,568) 2018-06-19 11:30
http://mantis.whdload.de/file_download.php?file_id=623&type=bug

Notes
(0005958)
StingRay   
2017-12-21 14:15   
I've applied interrupt fixes for the copper and vblank interrupt. Problem still present with the attached slave?
(0005959)
Wepl   
2017-12-21 14:31   
With the new Slave exact the same fault appears.
(0005960)
StingRay   
2017-12-21 14:54   
That is pretty strange as the interrupts are definitely fixed so it shouldn't fault with ChkInts. Can you attach the register log once the new slave fails again please? Unless it is 100% the same as old one but that should not be case.
(0005961)
Wepl   
2017-12-21 14:58   
attached.
the code above has changed but listing is too short to end of actual interrupt code...
(0005962)
Wepl   
2017-12-21 15:01   
$0fbb6860 move.w #$20,($dff09c)
$0fbb6868 movem.l (a7)+,d0-d7/a0-a6 ;$0fc2837c
$0fbb686c rte
doesn't look good?
(0006254)
StingRay   
2018-06-19 11:29   
It indeed didn't look good, offsets for interrupt fixes were wrong (relocation header wasn't taken into account). Try attached slave please, the patch should now not fault anymore with ChkInts.
(0006257)
Wepl   
2018-06-19 22:54   
works with ChkInts now
still some gfx trembling, probably a timing issue and hard to locate
(0006259)
StingRay   
2018-06-20 14:47   
Thanks for super fast testing. :) At least the interrupts are fixed now, I'll check the code more thoroughly in the coming days/weeks to find the cause of the trembling graphics.

Issue History
2017-12-21 12:50WeplNew Issue
2017-12-21 12:50WeplFile Added: blastar.whdl_register
2017-12-21 12:50WeplAssigned To => StingRay
2017-12-21 12:50WeplStatusnew => assigned
2017-12-21 12:51WeplDescription Updatedbug_revision_view_page.php?rev_id=910#r910
2017-12-21 14:15StingRayFile Added: Blastar.slave
2017-12-21 14:15StingRayNote Added: 0005958
2017-12-21 14:31WeplNote Added: 0005959
2017-12-21 14:54StingRayNote Added: 0005960
2017-12-21 14:58WeplFile Added: blaster.2.whdl_register
2017-12-21 14:58WeplNote Added: 0005961
2017-12-21 15:01WeplNote Added: 0005962
2018-06-19 11:29StingRayNote Added: 0006254
2018-06-19 11:30StingRayFile Added: Blastar-2.slave
2018-06-19 22:54WeplNote Added: 0006257
2018-06-20 14:47StingRayNote Added: 0006259
2019-06-24 11:42StingRayAssigned ToStingRay =>
2020-04-08 00:04JOTDAssigned To => JOTD