2005 g35 won't reflash

nwellinghoff · **Joined:** Thu Apr 15, 2021 12:50 am **Posts:** 4

Hello,

I was able to successfully download my ROM from my 2005 g35 AT and load it up in RR with the definitions from Pytrex. I disabled a the P0420 code and saved the image. When attempting to reflash I did the p option and get the error.

Code:

> p
reflashing selected blocks (dry run). Note, some (harmless) write verification errors WILL
occur if there are "modified blocks" ! (i.e. ROM file differs from ECU ROM)
        Block 07
Erasing block 7 (0x007000-0x007FFF)...
writing chunk @ 0x007B00 ( 68 %,  4286 B/s, ~    0 s remaining)
        Problem: bad response @ 7B00: post-write verify failed

When I look in the def file it identifies my ECU as a AC700 as does the part number lookup when I do it by VIN. However, in the def file it says that is a ECU for a 2004. Could the checksum locations be different or something? Any suggestions for how to proceed?

Here is what nisprog says when it connects

Code:

nisprog> npconn
Connected to ECU !
Using short headers.
ECUID: AC700
Key candidate   dist (smaller is better)
0: 0xEED9A107   0
1: 0xEC5B98DE   2
2: 0x968148AD   12

Pytrex · **Posted:** Wed Apr 28, 2021 10:01 am

Checksums are more than likely fine. If RomRaider didn’t pick up an error when opening the ROM, then it’s most likely accurate. The error is normal because you’re doing a test flash. You’re not actually reflashing the ROM. If you read the notice, it says that you will get a post write verify error, which is exactly what happened.

Next time, just type “y” to reflash the changed blocks and it’ll flash as intended.

Edit; Make sure to disable both P0420 and P0430.

fenugrec · **Joined:** Wed Jan 08, 2014 11:07 pm **Posts:** 652

nwellinghoff wrote:

When attempting to reflash I did the p option and get the error.

Read the 4-5 posts here
viewtopic.php?p=158820#p158820

nwellinghoff · **Joined:** Thu Apr 15, 2021 12:50 am **Posts:** 4

Ah, thanks. I was able to reflash after ignoring the error. The error at the end is a little scary for a n00b so that is why I took pause. Maybe if the write verification errors were noted and the error was softened it would save you from having to answer this question again

I am no C programmer and couldn't get the build to work with VS but maybe something that gives lets you know its ok and returns success.

Code:

np_backend.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/np_backend.c b/np_backend.c
index 83c6a70..995e6a5 100644
--- a/np_backend.c
+++ b/np_backend.c
@@ -719,8 +719,13 @@ static int npk_raw_flashblock(const uint8_t *src, uint32_t start, uint32_t len)
             errval = diag_l1_recv(global_l2_conn->diag_link->l2_dl0d, NULL, &rxbuf[errval], needed, 300);
          }
          if (errval < 0) errval = 0;   //floor
-         printf("%s\n", decode_nrc(&rxbuf[1]));
+         char *errstring = decode_nrc(&rxbuf[1]);
+         printf("%s\n", errstring);
          (void) diag_l2_ioctl(global_l2_conn, DIAG_IOCTL_IFLUSH, NULL);
+         //take note that the error was a post-write verify such that the error can be softened in practice mode
+         if (strcmp(errstring, "post-write verify failed") == 0 ) {
+            return -2;
+         }
          return -1;
       }
 
@@ -808,7 +813,10 @@ int reflash_block(const uint8_t *newdata, const struct flashdev_t *fdt, unsigned
 
    /* 4- write */
    errval = npk_raw_flashblock(newdata, start, len);
-   if (errval) {
+   if (practice && errval == -2){
+      printf("\n post-write verify failed errors are expected in practice mode as the ROM was not actually written.");
+   }
+   else if (errval) {
       printf("\nReflash error ! Do not panic, do not reset the ECU immediately. The kernel is "
             "most likely still running and receiving commands !\n");
       goto badexit;

RomRaider

Forum rules

2005 g35 won't reflash

Who is online