On Sun, 6 Jan 2002, Matt wrote: > With some help from Christoph, I've been investigating the CB1 stairs bug > (#326).
> I haven't done the stepping to see where the return values of kernel calls > diverges, but wanted to update the list on where I am with this bug. Stepping in both FSCI and SSCI has yielded these results: In a call to CanBeHere is where they diverge. pc=2575 acc=7924 o=0684 fp=6374 sp=637a prev=4 sbase=6336 globls=1958 &restmod=0 Step #3484570 2575: [B] callk CanBeHere[51] 04 Kernel params: (0684, 7924) pc=2578 acc=0001 o=0684 fp=6374 sp=6374 prev=4 sbase=6336 globls=1958 &restmod=0 Step #3484571 2578: [W] bnt 0033 [25ae] pc=257b acc=0001 o=0684 fp=6374 sp=6374 prev=4 sbase=6336 globls=1958 &restmod=0 Step #3484572 257b: [B] pTos 1c (signal) pc=257d acc=0001 o=0684 fp=6374 sp=6376 prev=4 sbase=6336 globls=1958 &restmod=0 Step #3484573 257d: [W] ldi 2000 pc=2580 acc=2000 o=0684 fp=6374 sp=6376 prev=4 sbase=6336 globls=1958 &restmod=0 Step #3484574 2580: [W] and pc=2581 acc=2000 o=0684 fp=6374 sp=6374 prev=4 sbase=6336 globls=1958 &restmod=0 Step #3484575 2581: [W] bt 0015 [2599] pc=2599 acc=2000 o=0684 fp=6374 sp=6374 prev=4 sbase=6336 globls=1958 &restmod=0 Step #3484576 2599: [W] bnt 0012 [25ae] pc=259c acc=2000 o=0684 fp=6374 sp=6374 prev=4 sbase=6336 globls=1958 &restmod=0 Step #3484577 259c: [B] pTos 48 (blocks) pc=259e acc=2000 o=0684 fp=6374 sp=6376 prev=4 sbase=6336 globls=1958 &restmod=0 Step #3484578 259e: [B] ldi 00 pc=25a0 acc=0000 o=0684 fp=6374 sp=6376 prev=4 sbase=6336 globls=1958 &restmod=0 Step #3484579 25a0: [W] eq? pc=25a1 acc=0001 o=0684 fp=6374 sp=6374 prev=0 sbase=6336 globls=1958 &restmod=0 Step #3484580 25a1: [W] bt 000a [25ae] pc=25ae acc=0001 o=0684 fp=6374 sp=6374 prev=0 sbase=6336 globls=1958 &restmod=0 Step #3484581 25ae: [W] ret -------- same up to this point ------- pc=776c acc=0000 o=aada fp=6360 sp=636c prev=0 sbase=6336 globls=1958 &restmod=0 Step #3484582 776c: [B] pTos 0c (x) ----- This is where they diverge. In SSCI, the accumulator is 1. in FSCI, it is 0. FSCI is one pixel off on the x coordinate, so that may affect this. It seems like I can't get ego into the same position in both FSCI and SSCI. Is there any way to change variables in FSCI or SSCI to "force" ego into a certain position? -- http://www.clock.org/~matt
