pcmhacking.net

Electronic Fuel Injection - Developement & Tuning
https://pcmhacking.net/forums/

PCM Hammer P12 development

https://pcmhacking.net/forums/viewtopic.php?t=7742

Page 10 of 56

Re: PCM Hammer fails on P12

Posted: Thu Feb 24, 2022 8:11 pm
by Gampy
The Build.cmd system included with the PCMHacks suite allows you to dump the intermediate assembly using option -d ... Just type,
Build.cmd -h
In the PCMHacks suite kernel directory to see all it's options! (the .cmd is important)

The first Empty routine (this one completed successfully) was,

Code: Select all

Disassembly of section .kernel_code:

00ff2000 <KernelStart>:
  ff2000:	4e75           	rts
	...

Disassembly of section .text:

12340000 <_start>:
12340000:	4e75           	rts
12340002:	0000 4743      	orib #67,%d0
12340006:	433a 2028      	chkl %pc@(12342030 <_start+0x2030>),%d1
1234000a:	474e           	.short 0x474e
1234000c:	5529 2034      	subqb #2,%a1@(8244)
12340010:	2e38 2e30      	movel 2e30 <KernelStart-0xfef1d0>,%d7
Test 3 out now ...

Code: Select all

Disassembly of section .kernel_code:

00ff2000 <KernelStart>:
  ff2000:	4eb9 00ff 2008 	jsr ff2008 <ScratchWatchdog>
  ff2006:	4e75           	rts

00ff2008 <ScratchWatchdog>:
  ff2008:	203c 00ff fa55 	movel #16775765,%d0
  ff200e:	2040           	moveal %d0,%a0
  ff2010:	10bc 0055      	moveb #85,%a0@
  ff2014:	203c 00ff fa55 	movel #16775765,%d0
  ff201a:	2040           	moveal %d0,%a0
  ff201c:	10bc ffaa      	moveb #-86,%a0@
  ff2020:	223c 00ff fa21 	movel #16775713,%d1
  ff2026:	203c 00ff fa21 	movel #16775713,%d0
  ff202c:	2040           	moveal %d0,%a0
  ff202e:	1010           	moveb %a0@,%d0
  ff2030:	0200 007f      	andib #127,%d0
  ff2034:	2041           	moveal %d1,%a0
  ff2036:	1080           	moveb %d0,%a0@
  ff2038:	223c 00ff fa21 	movel #16775713,%d1
  ff203e:	203c 00ff fa21 	movel #16775713,%d0
  ff2044:	2040           	moveal %d0,%a0
  ff2046:	1010           	moveb %a0@,%d0
  ff2048:	0000 ff80      	orib #-128,%d0
  ff204c:	2041           	moveal %d1,%a0
  ff204e:	1080           	moveb %d0,%a0@
  ff2050:	4e75           	rts

Disassembly of section .text:

12340000 <_start>:
12340000:	4e75           	rts
12340002:	0000 4743      	orib #67,%d0
12340006:	433a 2028      	chkl %pc@(12342030 <_start+0x2030>),%d1
1234000a:	474e           	.short 0x474e
1234000c:	5529 2034      	subqb #2,%a1@(8244)
12340010:	2e38 2e30      	movel 2e30 <KernelStart-0xfef1d0>,%d7
Test 4 out now ...

Code: Select all

Disassembly of section .kernel_code:

00ff2000 <KernelStart>:
  ff2000:	4eb9 00ff 2008 	jsr ff2008 <ScratchWatchdog>
  ff2006:	4e75           	rts

00ff2008 <ScratchWatchdog>:
  ff2008:	203c 00ff fa55 	movel #16775765,%d0
  ff200e:	2040           	moveal %d0,%a0
  ff2010:	10bc 0055      	moveb #85,%a0@
  ff2014:	203c 00ff fa55 	movel #16775765,%d0
  ff201a:	2040           	moveal %d0,%a0
  ff201c:	10bc ffaa      	moveb #-86,%a0@
  ff2020:	223c 00ff fa21 	movel #16775713,%d1
  ff2026:	203c 00ff fa21 	movel #16775713,%d0
  ff202c:	2040           	moveal %d0,%a0
  ff202e:	1010           	moveb %a0@,%d0
  ff2030:	0200 007f      	andib #127,%d0
  ff2034:	2041           	moveal %d1,%a0
  ff2036:	1080           	moveb %d0,%a0@
  ff2038:	4e71           	nop
  ff203a:	4e71           	nop
  ff203c:	4e71           	nop
  ff203e:	4e71           	nop
  ff2040:	4e71           	nop
  ff2042:	4e71           	nop
  ff2044:	223c 00ff fa21 	movel #16775713,%d1
  ff204a:	203c 00ff fa21 	movel #16775713,%d0
  ff2050:	2040           	moveal %d0,%a0
  ff2052:	1010           	moveb %a0@,%d0
  ff2054:	0000 ff80      	orib #-128,%d0
  ff2058:	2041           	moveal %d1,%a0
  ff205a:	1080           	moveb %d0,%a0@
  ff205c:	4e75           	rts

Disassembly of section .text:

12340000 <_start>:
12340000:	4e75           	rts
12340002:	0000 4743      	orib #67,%d0
12340006:	433a 2028      	chkl %pc@(12342030 <_start+0x2030>),%d1
1234000a:	474e           	.short 0x474e
1234000c:	5529 2034      	subqb #2,%a1@(8244)
12340010:	2e38 2e30      	movel 2e30 <KernelStart-0xfef1d0>,%d7
I'll work on the contributed thoughts ... And spend some time in the disassembly.

Re: PCM Hammer fails on P12

Posted: Fri Feb 25, 2022 12:03 am
by Gampy
Both last tests were successful, I've sent another with a delayed loop scratching the dog, that has a half dozen nop's in between COP2 bset and bclr, that should live for ~20 seconds.

Code: Select all

Disassembly of section .kernel_code:

00ff2000 <KernelStart>:
  ff2000:	598f           	subql #4,%sp
  ff2002:	4297           	clrl %sp@
  ff2004:	6008           	bras ff200e <KernelStart+0xe>
  ff2006:	4eb9 00ff 2078 	jsr ff2078 <LongSleepWithWatchdog>
  ff200c:	5297           	addql #1,%sp@
  ff200e:	7031           	moveq #49,%d0
  ff2010:	b097           	cmpl %sp@,%d0
  ff2012:	6cf2           	bges ff2006 <KernelStart+0x6>
  ff2014:	588f           	addql #4,%sp
  ff2016:	4e75           	rts

00ff2018 <WasteTime>:
  ff2018:	4e71           	nop
  ff201a:	4e71           	nop
  ff201c:	4e71           	nop
  ff201e:	4e71           	nop
  ff2020:	4e75           	rts

00ff2022 <ScratchWatchdog>:
  ff2022:	203c 00ff fa55 	movel #16775765,%d0
  ff2028:	2040           	moveal %d0,%a0
  ff202a:	10bc 0055      	moveb #85,%a0@
  ff202e:	203c 00ff fa55 	movel #16775765,%d0
  ff2034:	2040           	moveal %d0,%a0
  ff2036:	10bc ffaa      	moveb #-86,%a0@
  ff203a:	223c 00ff fa21 	movel #16775713,%d1
  ff2040:	203c 00ff fa21 	movel #16775713,%d0
  ff2046:	2040           	moveal %d0,%a0
  ff2048:	1010           	moveb %a0@,%d0
  ff204a:	0200 007f      	andib #127,%d0
  ff204e:	2041           	moveal %d1,%a0
  ff2050:	1080           	moveb %d0,%a0@
  ff2052:	4e71           	nop
  ff2054:	4e71           	nop
  ff2056:	4e71           	nop
  ff2058:	4e71           	nop
  ff205a:	4e71           	nop
  ff205c:	4e71           	nop
  ff205e:	223c 00ff fa21 	movel #16775713,%d1
  ff2064:	203c 00ff fa21 	movel #16775713,%d0
  ff206a:	2040           	moveal %d0,%a0
  ff206c:	1010           	moveb %a0@,%d0
  ff206e:	0000 ff80      	orib #-128,%d0
  ff2072:	2041           	moveal %d1,%a0
  ff2074:	1080           	moveb %d0,%a0@
  ff2076:	4e75           	rts

00ff2078 <LongSleepWithWatchdog>:
  ff2078:	518f           	subql #8,%sp
  ff207a:	42af 0004      	clrl %sp@(4)
  ff207e:	6018           	bras ff2098 <LongSleepWithWatchdog+0x20>
  ff2080:	4eba ffa0      	jsr %pc@(ff2022 <ScratchWatchdog>)
  ff2084:	4297           	clrl %sp@
  ff2086:	6006           	bras ff208e <LongSleepWithWatchdog+0x16>
  ff2088:	4eba ff8e      	jsr %pc@(ff2018 <WasteTime>)
  ff208c:	5297           	addql #1,%sp@
  ff208e:	7009           	moveq #9,%d0
  ff2090:	b097           	cmpl %sp@,%d0
  ff2092:	6cf4           	bges ff2088 <LongSleepWithWatchdog+0x10>
  ff2094:	52af 0004      	addql #1,%sp@(4)
  ff2098:	0caf 0000 270f 	cmpil #9999,%sp@(4)
  ff209e:	0004 
  ff20a0:	6fde           	bles ff2080 <LongSleepWithWatchdog+0x8>
  ff20a2:	508f           	addql #8,%sp
  ff20a4:	4e75           	rts

Disassembly of section .text:

12340000 <_start>:
12340000:	4e75           	rts
12340002:	0000 4743      	orib #67,%d0
12340006:	433a 2028      	chkl %pc@(12342030 <_start+0x2030>),%d1
1234000a:	474e           	.short 0x474e
1234000c:	5529 2034      	subqb #2,%a1@(8244)
12340010:	2e38 2e30      	movel 2e30 <KernelStart-0xfef1d0>,%d7
I wanna see this live so bad I'm going bananas, I just hope I don't slip on the peel ... :roll:

Re: PCM Hammer fails on P12

Posted: Fri Feb 25, 2022 12:54 am
by Gampy
Bummer, it did not live.

Going to have to add some complexity!

Edit;
kur4o,

That disassembly, what is it, what is it's purpose in life ... PM me if you want.

Re: PCM Hammer fails on P12

Posted: Fri Feb 25, 2022 9:33 am
by Tazzi
Bugger, I'll see if I cant get my hands on a GM write kernel and check if theres something not missing.

Re: PCM Hammer fails on P12

Posted: Fri Feb 25, 2022 10:22 am
by darkman5001
Guys, I finally got my chip adapters and was able to pull the flash. Let me know if any of you are able to read it.

Re: PCM Hammer fails on P12

Posted: Fri Feb 25, 2022 12:10 pm
by Gampy
Awesome!

Might I make a suggestion ... The filename, add the OsID and family (P12) to it, it's reported as: 12606400

Re: PCM Hammer fails on P12

Posted: Fri Feb 25, 2022 12:16 pm
by darkman5001
Gampy wrote:Awesome!

Might I make a suggestion ... The filename, add the OsID to it, it's reported as: 12606400

Good idea. Were you able you read it at all?

Re: PCM Hammer fails on P12

Posted: Fri Feb 25, 2022 12:29 pm
by antus
Thanks! The file looks correct. Can see the right file structure and it looks like all data lines and address lines were well connected and the data does not look damaged.
It did need 'a' and 'b' bytes swapped. Ive applied this and now it loads correctly in ida.

Re: PCM Hammer fails on P12

Posted: Fri Feb 25, 2022 12:45 pm
by darkman5001
Greetings Antus. I am still trying to figure out how I might read the bin of the flash that I saved. How can I read it?

Re: PCM Hammer fails on P12

Posted: Fri Feb 25, 2022 1:13 pm
by Gampy
antus wrote:It did need 'a' and 'b' bytes swapped. Ive applied this and now it loads correctly in ida.
The m68k is big endian ... I thought IDA did the byte swap, maybe that is why my m68k disassemblies turn out like crap.
darkman5001 wrote:Greetings Antus. I am still trying to figure out how I might read the bin of the flash that I saved. How can I read it?
If you're talking about disassembling it, most use either IDA Pro or Ghidra.
All times are UTC+10:00
Page 10 of 56

Powered by phpBB® Forum Software © phpBB Limited