SAD Version 5 (alpha)

[tvrfan]

Thanks for KIEE will put that in my test set.

Yes, not fixed 'if (0 > 0) ' etc yet... on my list to do. Also may put back the 'gap scanner' for missing code.
I wanted to get the basic analysis correct as much as possible before putting more guesswork back (which is what it is, kind of).
V4 produced some false matches so I took the scanners out.

Names -
I have changed stuff over time to match the Ford manuals, which did appear publicly after I started SAD.
I don't mind which way round, I chose the Ford version to start with, but whatever everyone prefers... same for Special Func Registers - currently uses the handbook names, but can change to suit. Some 8065 may still be wrong as it can have several different modes for output etc. And I think it's set by hardware, so no (easy) way to tell from the binary.

Note - 5.0.7 does better on missed arguments after a fix, but M0M2 still not right.

[jsa]

Thanks for KIEE will put that in my test set.

I wanted to get the basic analysis correct as much as possible

Thank you for adding KIEE to your test suite.

For code basics I'm seeing lots of code in all banks overlooked in RZASAMY with 5.07.

Code: Select all

     # Call  from L1B3F7                                                           ?1B358?
     # Call  from L1B565                                                           ?1B563?
     # Call  from L1B6D8                                                           ?1B6CE?
     # Call  from L1B71C                                                           ?1B6EF?
     # Call  from L1B78F                                                           ?1B76C?
     # Call  from L1B8D8                                                           ?1B8CB?
     # Call  from L1B975                                                           ?1B96A?
     # Call  from L1BAEC                                                           ?1BAC7?
     # Call  from L1BB16                                                           ?1BAC7?
     # Call  from L1BB30                                                           ?1BAC7?
     # Call  from L1C459                                                           ?1C30E?
     # Call  from L1C69D                                                           ?1C635?
021b6: f2,71,df,36,71,bf  ???   
021bc: 36,a1,66,96,46,b1  ???   
021c2: ff,3e,c7,01,04,f4  ???   
021c8: 3e,c7,d4,7b,00,c7  ???   
021ce: 01,05,f4,00,b1,0a  ???   
021d4: 44,c4,3e,13,c0,3c  ???   

     # JNE   from L021DE                                                           ?021B6?
021da: 06,98,3e,13,d7,f5  ???   
021e0: 0c,04,3c,67,ce,2e  ???   
021e6: 3c,c3,d4,76,3c,71  ???   

     # SJump from L02319                                                           ?021B6?
021ec: fe,36,fa,29,4b,47  ???   
021f2: ce,2c,06,32,b2,46  ???   
021f8: 28,2a,8a,3f,0a,fd  ???   

     # JB    from L021FB HSO_BUFF_FULL = 1                                         ?021B6?
021fe: 11,1f,a0,32,0e,b1  ???   
02204: 2a,0d,07,46,89,69  ???   
0220a: 96,46,d3,04,a1,66  ???   
02210: 96,46,a0,06,3c,08  ???   

     # JNC   from L0220C                                                           ?021B6?
02216: 01,3c,47,ce,08,3c  ???   
0221c: 24,01,40,11,2c,b1  ???   

     # SJump from L022DB                                                           ?021B6?
02222: 01,28,17,05,a0,06  ???   

     # JB    from L02239 B1_TMP1H = 1                                              ?021B6?
     # SJump from L0225E                                                           ?021B6?
02228: 3c,08,01,3c,68,24  ???   
0222e: 3c,3f,3d,02,20,b1  ???   

     # JB    from L0222F B7_TMP5H = 1                                              ?021B6?
02234: b3,2f,8c,00,35,39  ???   
0223a: 35,e8,47,ce,04,06  ???   
02240: 3c,b3,cf,b6,00,31  ???   
02246: b3,cf,b7,00,3e,b0  ???   

     # JNB   from L02258 B1_TMP1H = 0                                              ?021B6?
0224c: 3e,30,b3,2f,8c,00  ???   

     # DJNZ  from L0225B                                                           ?021B6?
02252: 35,e0,31,02,20,08  ???   

     # DJNZ  from L02253                                                           ?021B6?
02258: 31,35,f0,e0,30,f0  ???   
0225e: 27,c4,a3,ce,02,30  ???   

     # SJump from L02256                                                           ?021B6?
     # JNC   from L022A1                                                           ?021B6?

     # JB    from L02268 B7_TMP6H = 1                                              ?021B6?
02264: 48,3c,06,3e,3f,3f  ???   
0226a: f9,b3,2f,8c,00,35  ???   

     # DJNZ  from L02275                                                           ?021B6?
02270: 31,35,02,17,30,e0  ???   

     # JNB   from L02270 B1_TMP1H = 0                                              ?021B6?
02276: 31,f3,11,2b,9a,ce  ???   
0227c: 30,d1,05,90,28,2c  ???   
02282: 17,2b,99,01,40,d1  ???   

     # JLEU  from L0227D                                                           ?021B6?
02288: 11,08,01,3a,d3,05  ???   
0228e: 38,2b,09,20,03,30  ???   

     # JNC   from L0228C                                                           ?021B6?
02294: 2b,04,85,08,84,3a  ???   

     # SJump from L02291                                                           ?021B6?

     # JLEU  from L02287                                                           ?021B6?
     # JB    from L0228E B0_TEMP3H = 1                                             ?021B6?
     # JNB   from L02293 B0_TEMP3H = 0                                             ?021B6?
0229a: 67,ce,06,3c,19,01  ???   
022a0: 28,d3,bd,c7,41,04  ???   
022a6: f4,2c,99,01,40,d9  ???   
022ac: 25,df,1f,34,2d,3b  ???   

     # JE    from L022EB                                                           ?021B6?
022b2: 47,ce,1e,06,32,28  ???   
022b8: 83,28,db,29,ee,a0  ???   
022be: 32,3c,08,01,32,47  ???   
022c4: ce,0a,32,24,a1,91  ???   
022ca: 37,3a,20,04,67,ce  ???   

     # JE    from L022AD                                                           ?021B6?
022d0: 0c,24,07,40,28,5c  ???   

     # JGTU  from L022AB                                                           ?021B6?
     # SJump from L022CC                                                           ?021B6?
022d6: 99,0b,40,df,02,27  ???   
022dc: 42,88,00,3a,df,39  ???   

     # JE    from L022D9                                                           ?021B6?
022e2: 91,01,36,34,2d,05  ???   

     # SJump from L02232                                                           ?021B6?
022e8: 88,00,40,df,c5,3f  ???   

     # JNB   from L022AF B4_TEMP4H = 0                                             ?021B6?
     # JNB   from L022E5 B4_TEMP4H = 0                                             ?021B6?
     # JB    from L022ED HSO_BUFF_FULL = 1                                         ?021B6?
022ee: 0a,fd,11,1f,45,03  ???   
022f4: 00,06,0e,b1,0a,0d  ???   
022fa: fb,34,2d,2d,28,2d  ???   
02300: 15,44,df,27,c4,3e  ???   

     # JNE   from L0230D                                                           ?021B6?
02306: 13,c0,3c,06,98,3e  ???   
0230c: 13,d7,f5,0c,04,3c  ???   
02312: 6b,d4,76,3c,37,3d  ???   
02318: 12,26,d0,28,1f,fb  ???   

     # JE    from L022E0                                                           ?021B6?
0231e: 91,40,36,28,0a,3f  ???   
02324: 45,05,10,01,ef,83  ???   
0232a: 98,f3,f0,47,ce,22  ???   

     # JNB   from L022FB B4_TEMP4H = 0                                             ?021B6?
     # JE    from L02302                                                           ?021B6?
     # JNB   from L02316 B7_TMP5H = 0                                              ?021B6?
     # JB    from L02323 B7_TMP9H = 1                                              ?021B6?

     # SCall from L022FE                                                           ?021B6?
     # SCall from L02321                                                           ?021B6?
02330: 06,3c,17,05,48,3c  ???   

     # SCall from L022D4                                                           ?021B6?
     # JB    from L02338 B7_TMP6H = 1                                              ?02332?
     # SCall from L02479                                                           ?02436?
     # SCall from L024A1                                                           ?02485?
02336: 06,3e,3f,3f,f7,f0  ???   

     # SCall from L021EF                                                           ?021B6?
     # SCall from L022B7                                                           ?021B6?
     # SCall from L0231B                                                           ?021B6?
     # Call  from L1BE8D                                                           ?1BE87?
0233c: f2,b3,2f,8a,00,3e  ???   
02342: 38,3e,1c,c7,2e,3d  ???   
02348: 00,a1,f2,2b,3e,10  ???   
0234e: 01,ef,9c,9b,c3,2e  ???   
02354: 3c,3e,10,01,ef,93  ???   
0235a: 9b,c3,2e,3c,3e,20  ???   
02360: 09,91,08,b6,ef,f9  ???   

     # JB    from L02342 B0_TMP6L = 1                                              ?0233C?
02366: 04,71,f7,b6,a1,40  ???   

     # SJump from L0235F                                                           ?0233C?
0236c: 0f,42,a1,ed,cb,3c  ???   
02372: c7,42,22,3c,c7,42  ???   
02378: 2e,3d,a1,ff,ff,3e  ???   
0237e: c3,42,21,3e,c7,01  ???   
02384: 62,0f,3c,c7,01,6e  ???   
0238a: 0f,3d,b1,41,42,c3  ???   
02390: 01,6a,0f,42,f3,f0  ???   

     # SCall from L022B9                                                           ?021B6?
02396: a3,da,78,24,a1,15  ???   
0239c: 46,28,10,01,ef,a4  ???   
023a2: 94,c3,da,78,24,a3  ???   
023a8: da,7a,24,a1,2d,57  ???   
023ae: 28,10,01,ef,93,94  ???   
023b4: c3,da,7a,24,a3,da  ???   
023ba: 7c,24,a1,dd,be,28  ???   
023c0: 10,01,ef,82,94,c3  ???   
023c6: da,7c,24,10,01,ef  ???   
023cc: 2c,a3,b1,03,2c,b3  ???   
023d2: 01,02,f4,c4,3e,c4  ???   
023d8: 19,3f,45,29,b1,13  ???   
023de: 28,c7,da,77,2c,a1  ???   
023e4: f7,08,26,b1,07,29  ???   
023ea: 28,4a,67,ce,20,32  ???   
023f0: 20,36,b1,0f,28,b1  ???   

     # JB    from L023D6 B6_FLAGTEMP_BG = 1                                        ?02396?
023f6: 03,2c,a1,2c,00,26  ???   
023fc: b1,02,29,28,35,28  ???   
02402: 26,20,23,b1,11,28  ???   

     # JB    from L023D9 B7_TMP9H = 1                                              ?02396?
02408: c7,01,2f,f4,2c,a1  ???   
0240e: 2f,f4,26,32,2d,09  ???   
02414: c7,01,37,f4,2c,a1  ???   
0241a: 37,f4,26,b1,07,29  ???   

     # JNB   from L02411 B2_TEMP4H = 0                                             ?02396?
02420: 28,14,28,05,67,ce  ???   
02426: 1c,32,f0,b0,3a,28  ???   

     # SJump from L023F0                                                           ?02396?
     # SJump from L02403                                                           ?02396?

     # SCall from L02401                                                           ?02396?
     # SCall from L02422                                                           ?02396?
0242c: b0,3b,2c,a1,2c,00  ???   
02432: 26,b1,02,29,a1,91  ???   

     # SCall from L023EA                                                           ?02396?
     # SCall from L023FF                                                           ?02396?
     # SCall from L02420                                                           ?02396?
02438: 37,3a,b1,08,2a,28  ???   

     # DJNZ  from L02481                                                           ?02436?

     # DJNZ  from L0247B                                                           ?02436?
0243e: 6c,47,ce,28,32,3c  ???   
02444: a3,ce,2a,3e,11,2b  ???   
0244a: 18,01,28,d3,0b,47  ???   
02450: ce,24,32,3c,a3,ce  ???   
02456: 26,3e,17,2b,3f,0a  ???   

     # JNC   from L0244D                                                           ?02436?
     # JB    from L0245A HSO_BUFF_FULL = 1                                         ?02436?
0245c: fd,11,1f,a0,3c,0e  ???   
02462: b1,2a,0d,64,3e,32  ???   
02468: 08,01,3a,d3,05,38  ???   
0246e: 2b,09,20,03,30,2b  ???   

     # JNC   from L0246B                                                           ?02436?
02474: 04,85,08,84,3a,2e  ???   

     # SJump from L02470                                                           ?02436?

     # JB    from L0246D B0_TEMP3H = 1                                             ?02436?
     # JNB   from L02472 B0_TEMP3H = 0                                             ?02436?
0247a: b7,e0,2a,bf,b2,27  ???   
02480: 28,e0,29,b6,f0,b1  ???   

     # SCall from L021F9                                                           ?021B6?
02486: 08,2a,b1,2a,2b,18  ???   

     # DJNZ  from L024A7                                                           ?02485?
0248c: 01,28,d3,03,b1,0a  ???   
02492: 2b,3f,0a,fd,11,1f  ???   

     # JNC   from L0248E                                                           ?02485?
     # JB    from L02493 HSO_BUFF_FULL = 1                                         ?02485?
02498: a0,32,0e,b0,2b,0d  ???   
0249e: a0,32,3c,2e,8f,67  ???   
024a4: ce,2c,32,e0,2a,de  ???   
024aa: f0,3f,0a,fd,11,1f  ???   

     # SCall from L022BB                                                           ?021B6?
     # SCall from L0243D                                                           ?02436?
     # JB    from L024AB HSO_BUFF_FULL = 1                                         ?024AB?
024b0: a0,32,0e,b1,0a,0d  ???   
024b6: f0                 ???   


     # JNB   from L024C4 DATA = 0                                                  ?024AB?
024b7: 09,0b,00           shlw  R0,b             ZERO <<= b;
024ba: 09,03,00           shlw  R0,3             ZERO <<= 3;
024bd: 3a,0a,07           jb    B2,Ra,024c7      if (DATA = 0) {

Also seeing a lot of Bank 3 gremlins.

Code: Select all

     # JNE   from L02CDB                                                           Sub_0297b
02cca: b3,4d,60,99,04     ldb   R4,[R4c+9960]    AD_Cmd = [FGTMP2L+39960];

     # JNB   from L02CCF ATOD = 0                                                  B0_Sub_0297b
02ccf: 33,0a,fd           jnb   B3,Ra,02ccf      if (ATOD = 0) goto 02ccf;
02cd2: 51,0f,04,49        an3b  R49,R4,f         FGTMP0H = ATOD_LOW & f;
02cd6: 9b,4d,60,99,49     cmpb  R49,[R4c+9960]   
02cdb: d7,ed              jne   02cca            if (FGTMP0H != [FGTMP2L+39960]) goto 02cca;
02cdd: 51,f0,04,4e        an3b  R4e,R4,f0        FGTMP3L = ATOD_LOW & f0;
02ce1: b0,05,4f           ldb   R4f,R5           FGTMP3H = ATOD_HI;
02ce4: c3,4d,20,04,4e     stw   R4e,[R4c+420]    [FGTMP2L+30420] = FGTMP3L;
02ce9: c3,4b,00,04,4e     stw   R4e,[R4a+400]    [FGTMP1L+30400] = FGTMP3L;
02cee: 20,49              sjmp  02d39            goto 02d39; }


     # JE    from L02CC8                                                           Sub_0297b
02cf0: af,f0,02,52        ldzbw R52,[Rf0+2]      FGTMP5L = AICE_?x2;
02cf4: 08,01,52           shrw  R52,1            FGTMP5L >>= 1;

     # JNE   from L02D2D                                                           Sub_0297b
02cf7: b3,4d,5c,99,4e     ldb   R4e,[R4c+995c]   FGTMP3L = [FGTMP2L+3995c];

[BOOSTEDEVERYTHING]

The choice of V instead OVF, is intended to align SAD with the Manuals and Handbook.
When SAD was first released the manuals and handbook were not publicly available, so OVF was chosen on other criteria.
Now, the user can search the manuals and find PSW and the V flag to understand the meaning.

Ah, ok. So... The PSW_V is replacing the OVF? So the old version of SAD was definitely wrong in the screen shot I posted above? Just trying to understand. Sorry if this is in the wrong post.

[BOOSTEDEVERYTHING]

Names -
I have changed stuff over time to match the Ford manuals, which did appear publicly after I started SAD.
I don't mind which way round, I chose the Ford version to start with, but whatever everyone prefers... same for Special Func Registers - currently uses the handbook names, but can change to suit. Some 8065 may still be wrong as it can have several different modes for output etc. And I think it's set by hardware, so no (easy) way to tell from the binary.

As someone who is completely new to all this, I like the names matching the handbook and other posted docs. It makes it easier to search for something if I don't understand it.

[tvrfan]

As someone who is completely new to all this, I like the names matching the handbook and other posted docs. It makes it easier to search for something if I don't understand it.

Yes, that was the plan - use the Ford names to make them match in the manuals.

Overflow versus operands (PSW_V) - It wasn't always wrong in listings, just enough times to make me ditch it for now. Better than misleading someone.

[jsa]

Ah, ok. So... The PSW_V is replacing the OVF? So the old version of SAD was definitely wrong in the screen shot I posted above? Just trying to understand. Sorry if this is in the wrong post.

Yes, PSW_V replaces OVF.

Code: Select all

030df: 68,52,48           sb2w  R48,R52          FGTMP0L -= FGTMP5L;
030e2: d5,08              jnv   030ec            if (FGTMP0L >= 0) {

Jump Not V Set or Continue V Set.

If FGTMPOL changed from d32768 to d32767, PSW_V would be set, so continue.
In this case it is correct.

If FGTMPOL changed from d32767 to d32766, PSW_V would be clear, so jump.
In this case it is wrong.

[Jamie from oz]

Andy i think sad 5.0.7 may be not displaying added Rbase R22 in my 3DBC LST.

(3DBC) the other Rbase are commented out as they are only there for my reference.

5.0.6 dir

Code: Select all

### Symbol list ###

#rbase 72 46e  
#rbase a2 37c 
#rbase bc 180 
#rbase be 280 
rbase 22 0437 2455 251c

sym 2  "CPU_OK" :B 6                  # auto
sym 2  "LSO_Port"                     # auto

5.0.6 lst

Code: Select all

   Sub_2445_kam_qualification_test_logic:
2445: a1,37,04,22         ldw   R22,437          R22 = 437_?kam_bit_flags?;
2449: 71,7f,28            an2b  R28,7f           FlgR28_B7_kamok = 0;
244c: b1,aa,42            ldb   R42,aa           R42 = aa;
244f: b1,c6,30            ldb   R30,c6           Scr30 = c6;
2452: b1,75,32            ldb   R32,75           R32 = 75;
2455: 9b,22,b7,42         cmpb  R42,[R22+b7]     
2459: d7,2f               jne   248a             if (R42 = 3ee_KAMQA) {            # [r22-49] 437-49=3EE ? kamqa?   LOWEST ADDRESS OF KAM # KAMQA aa= 10101010 BINARY
245b: 9b,22,27,30         cmpb  R30,[R22+27]     
245f: d7,29               jne   248a             if (Scr30 = 45e_KAMQB) {          # [r22+27] 437+27=45E ? kamqb?   MIDDLE ADDRESS OF KAM # KAMQB c6= 11000110 BINARY
2461: 9b,22,50,32         cmpb  R32,[R22+50]     
2465: d7,23               jne   248a             if (R32 = 487_KAMQC) {   

sad 5.0.7 dir
i tried using the Rbase over a range like i had previously but all 3 dont appear to be working for me.
tried each individual over address range befor commenting them out and trying Rbase 22 0437.

Code: Select all

### Symbol list ###

rbase 22 0437
#rbase 22 0437 2455 251c
#rbase 22 0437 8504 8623
#rbase 72 46e  
#rbase a2 37c 
#rbase bc 180 
#rbase be 280 

sad 5.0.7 lst

Code: Select all

   Sub_2445_kam_qualification_test_logic:
2445: a1,37,04,22         ldw   R22,437          R22 = 437_?kam_bit_flags?;
2449: 71,7f,28            an2b  R28,7f           FlgR28_B7_kamok = 0;
244c: b1,aa,42            ldb   R42,aa           R42 = aa;
244f: b1,c6,30            ldb   R30,c6           Scr30 = c6;
2452: b1,75,32            ldb   R32,75           R32 = 75;
2455: 9b,22,b7,42         cmpb  R42,[R22+b7]     
2459: d7,2f               jne   248a             if (R42 = [R22-49]) {             # [r22-49] 437-49=3EE ? kamqa?   LOWEST ADDRESS OF KAM # KAMQA aa= 10101010 BINARY
245b: 9b,22,27,30         cmpb  R30,[R22+27]     
245f: d7,29               jne   248a             if (Scr30 = [R22+27]) {           # [r22+27] 437+27=45E ? kamqb?   MIDDLE ADDRESS OF KAM # KAMQB c6= 11000110 BINARY
2461: 9b,22,50,32         cmpb  R32,[R22+50]     
2465: d7,23               jne   248a             if (R32 = [R22+50]) {   

[tvrfan]

Thanks jamie - I did have a problem with 'phantom' rbases, and then after changing the code, Rbases don't even show up.....from too many to too little....
OK. more work to do.

3DBC OK. will check.

[Jamie from oz]

[quote=tvrfan post_id=13567

3DBC OK. will check.
[/quote]

If you don't have it I had posted it on eectuning a while back.

[tvrfan]

Released version 5.0.8 with some more fixes.

PROBLEM -
I have found things which mess up my 'elegant' v5 design, and I think it means some of my v5 rewrite won't ever work right... Damn, Damn, Damn!!! (and 'F' words) It may take a while to fix some of these issues. Very frustrating as I thought I had a neat solution, only to find it don't work for some bins. In the meantime v4 works OK, and you can use its output to feed into v5 to add calcs and new features, modify and/or make your own commands to fix the v5 issues.

So I may be quieter for a while...but I am still working on getting SAD to work as well as I can.