Re: [hatari-devel] WinUAE CPU core disassembler output options : fix for DISASM_FLAG_EA

[ Thread Index | Date Index | More lists.tuxfamily.org/hatari-devel Archives ]


Hi,

After applying your flags.patch, there are still other places showing (not any more relevant) values in CPU core disassembly:
------------------------------------
00E503B6 2470 C800       movea.l (a0,a4.l,$00) == $00e502ac,a2
....
00E506C2 4232 0000       clr.b (a2,d0.w,$00) == $00e729f8
....
00E50B04 43F2 0800       lea.l (a2,d0.l,$00) == $00e729f8,a1
....
00E50EC2 00EA 00F0 00F6  [ cmp2.b (a2,$00F6),d0 ]
00E50EC4 00F0 00F6 00FC           [ cmp2.b (a0,d0.w,$fc) == $00002300,d0 ]
00E50EC6 00F6 00FC 0102  [ cmp2.b (a6,d0.w,$02) == $0000f7ea (68020+),d0 ]
....
00E50EF0 0174 017A       bchg.b d0,(a4,d0.w,$7a) == $00e5262a (68020+)
....
00E50F06 01B6 01BC       bclr.b d0,(a6,d0.w,$bc) == $0000f7a4 (68020+)
------------------------------------

(And what's with 'cmp2' being in square brackets?)


	- Eero


On 26.11.2022 14.48, Nicolas Pomarède wrote:
Le 21/11/2022 à 02:26, Eero Tamminen a écrit :
Hi,

On 20.11.2022 23.42, Nicolas Pomarède wrote:
Le 20/11/2022 à 21:15, Eero Tamminen a écrit :
On 20.11.2022 21.23, Nicolas Pomarède wrote:
pushed the changes to WinUAE and they're now in Hatari, so disasm flags should work as expected now.

I'm still seeing these:
-------------------------
2470 C800  movea.l (a0,a4.l,$00) == $00e502ac,a2
...
206F 0004  movea.l (a7,$0004),a0
...
2070 0000  movea.l (a0,d0.w,$00) == $00002304,a0
...
4232 0000  clr.b (a2,d0.w,$00) == $00e729f8
-------------------------

So I guess some additional fixes are needed.

After that, having instruction and its parameters in separate columns (like with ext disassembler) would be final readability improvement.

(Having that "== ..." info in its own column would help readability when that extra info is configured to be shown.)

what DISASM_FLAG do you think is not correctly applied ?

By default Hatari disables all of these:
DISASM_FLAG_CC | DISASM_FLAG_EA | DISASM_FLAG_VAL

I think they should be enough to disable all value showing.

Here's a small patch to fix these missing checks of the flags

Toni, I'm not sure if the flag to test in these cases should be DISASM_FLAG_VAL or DISASM_FLAG_EA (the later seems to be used only in bsr/jsr and the like with pc relative values ?)

Nicolas







Mail converted by MHonArc 2.6.19+ http://listengine.tuxfamily.org/