RomRaider

Didnt know where else to put this reply, I still need to go and get codes for what is going on.

I extended my obd2 cable because the location is annoying and hits my leg while logging.

I purchased one of these.

https://www.jaycar.com.au/1m-obd-ii-extension-cable/p/LA9037

For some reason now nisprog is having a dummyspit. When i disconnect the extension cable i can connect so it isn't any other issue there. However what is likely to be occurring here? Im guessing its a timing issue but its only an extra 1m of wire can that make such a difference?

I don't know what that means.
If it's having problems after starting the kernel, it could be due to the higher speed (62.5kbps with the kernel, vs 10.4kbps for regular comms). The added capacitance of the cable slows down the edges, especially the rising edge. Depending on the design of the transceiver in your usb-obd adapter, this could be enough to cause issues. Different cables will behave differently.

If you're having trouble even with the initial connection, there's not much I can suggest. Your obd adpater should have a 510ohm resistor between K and V+ (battery voltage); if they used a higher resistance that could also explain it.

WIthout seeing an oscilloscope capture, all of the above is pure speculation.

I'm having the same issue right now but mine doesnt say 107 its 60507 or something.

Got it to work I swapped the 18 bin for 35.

Hello

I have done an slow dump to get this file:
https://www.romraider.com/forum/viewtopic.php?p=165297#p165297

It seems that keys are 0xFB518698 and 0x0BB956D3.
But it is a 384 kb ROM from SH7054 CPU.

Is it safe to flash a mod file with nisprog?

Thanks

Interesting. We haven't seen many 7054 dumps.

No, there is currently no kernel for 7052/7054, so it's not possible to reflash.

New release : nisprog 1.05

See first post for download links.

****** 1.05-2523639_20220616

-Support reflashing Subaru 7055 / 7058 over Kline , contributed by rimwall. Status is Alpha, experimental.
-added a bunch of ecukeys
-bump freediag for misc CLI fixes
-note : kernels are now in kernels/ instead of npkern/. You may need to update your .ini files

@fenugrec

I've checked out the GIT project (and associated subprojects) and required some suggested fixes to get this working with building under MINGW

cli_utils/md5/md5.c
-#include <md5.h>
+#include "md5.h"

cli_utils/nislib.c
void sum32(const uint8_t *buf, u32 siz, uint32_t *sum, uint32_t *xor)

Rename xor to something else (eg. xornum) since xor is a reserved token in later C versions

Also warning:
nisrom_keyfinders.c: In function 'keysearch_literal_cb':
nisrom_keyfinders.c:111:19: warning: 'needle' may be used uninitialized [-Wmaybe-uninitialized]
111 | bool rv = find_key_literal(lss->buf, lss->siz, needle, lss->thorough);

Thanks. If you're on github anyway, I prefer people to file issues there instead : easier to keep track, than misc posts sprinkled around a forum...



What compiler are you using ? I wanted to avoid modifying the md5 lib to keep it identical to its source (libmd) but will think about options.


Can you elaborate ? Only reference I see from C99 to C23 drafts is in <iso646.h> . Not that it's a problem to change but I want to know what complains about this...




That one is an annoying false positive - some analysis tools don't understand that assert(0) interrupts code flow, so this warning is spurious.

Let me know more about your build sys and workflow (preferably on github) and I'll see what I can do. Thanks for the comments.

I've put it up on Github. Version : g++.exe (Rev4, Built by MSYS2 project) 12.2.0

First error also with gcc (12.2.0) but XOR one only with later C++ versions. I'm working on decoding later versions of DAT in C++ rather C

Hello. Please help me figure out why kernel 7051, nissan maxima 2.0 is not loading. What could be the problem?

I just read the Nissan Cefiro SH 7051 ECU using this ini file #This is just an example .ini file to show some of the possible commands.
set
#Customize this to your hardware
port \\.\COM9

#This should always be OK
dumbopts 0x48

# No need to touch these unless you know what they are
l2protocol iso14230
initmode fast
testerid 0xfc
destaddr 0x10
addrtype phys

# now connect
up
nc

setdev 0
runkernel C:\nisprog\kernels\npk_SH7051.bin

up

help

##Enable read/write debugging output
#debug l1 0x8c


##watchmode
#watch 0xffff90bc


##Reduce delay between requests (faster dumps)
#npconf p3 0

##Extend read timeouts (in ms) (if you're getting read errors)
#npconf rxe 40

##connect
#nc

#Note, the old "setdev 0" etc. syntax is deprecated. Try "setdev ?" to show supported devices
#setdev 7055

##guess keys, if the suggested keyset doesn't work
#gk

#runkernel D:\dev\nis_kernels\npkern\npkern.bin

#stopkernel
#npdisc
#quit

10 min before Im read Cefiro ECU w this ini. But used not correct sidkey ECU read OK. Log attached

Nao finished reading. 2 dumps no differents

@ Yury123

The error is literally in the screenshot you posted. "Cannot open .../npk_sh7051" . You specified a bad path


@ Micharl2831

There are multiple errors, did you not read them ?

1-


setting proper device will be necessary for reflash. Also convenient for dumping , means you don't need to specify ROM size since nisprog will know.

2-



You formatted the setkeys command wrong. So of course SID27 fails and it falls back to the "slow" dump method.