Fred's first MTX tuning session!

MegaTunix is a cross-platform tuning application written and maintained by David J. Andruczyk. Discuss all things MTX here!
Post Reply
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Fred's first MTX tuning session!

Post by Fred »

This thread is to document the findings of a single half hour session driving around in the rain trying to sort the car out a bit better.

Firstly, thanks dave, MTX was a pleasure to use, mostly! :-)

X11 on the mac, freezing, not such a pleasure (3 times, had to reboot machine...).

Console afterwards had this in it, I'm going to try to find the binary logs now (on by default I hope, if not, fuck... oh well, next time...) to run through my parser and compare with this output. I took a few logs, hopefully MTX writes out the "custom" ones in real time such that when the machine dies you get whatever you get. Yep, one input log of 12megs and 20 minutes duration ctime/mtime. Got that, will run it through my parser now and paste results.

Code: Select all

FreeAir:bin fred$ ./a.out /Users/fred/first.tuning.session/inputlog-2011-03-05_2135.bin 
The length of file /Users/fred/first.tuning.session/inputlog-2011-03-05_2135.bin is 12119661

Attempting to parse file...

Packet number 11425 ending of length 81 at char number 1217502 failed checksum! Received 169 Calculated 199
Packet number 13228 ending of length 81 at char number 1407476 failed checksum! Received 95 Calculated 43
Packet number 20748 ending of length 82 at char number 2199613 failed checksum! Received 233 Calculated 203
Packet number 22088 ending of length 75 at char number 2340590 failed checksum! Received 157 Calculated 228
Packet number 24396 ending of length 186 at char number 2583443 failed checksum! Received 188 Calculated 83
Packet number 28047 ending of length 74 at char number 2968051 failed checksum! Received 200 Calculated 47
Packet number 34458 ending of length 81 at char number 3641678 failed checksum! Received 121 Calculated 127
Packet number 40035 ending of length 187 at char number 4227664 failed checksum! Received 27 Calculated 221
Packet number 44545 ending of length 70 at char number 4702466 failed checksum! Received 220 Calculated 128
Packet number 46189 ending of length 81 at char number 4875753 failed checksum! Received 35 Calculated 155
Packet number 47486 ending of length 72 at char number 5012176 failed checksum! Received 62 Calculated 201
Packet number 49366 ending of length 81 at char number 5209840 failed checksum! Received 226 Calculated 198
Packet number 60276 ending of length 82 at char number 6358648 failed checksum! Received 33 Calculated 186
Packet number 66225 ending of length 181 at char number 6985892 failed checksum! Received 116 Calculated 156
Packet number 67089 ending of length 71 at char number 7076813 failed checksum! Received 68 Calculated 107
Packet number 68687 ending of length 82 at char number 7244854 failed checksum! Received 168 Calculated 41
Packet number 69890 ending of length 81 at char number 7371601 failed checksum! Received 102 Calculated 110
Packet number 83477 ending of length 78 at char number 8801037 failed checksum! Received 88 Calculated 73
Packet number 84425 ending of length 102 at char number 8900744 failed checksum! Received 79 Calculated 190
Packet number 85041 ending of length 82 at char number 8965547 failed checksum! Received 132 Calculated 111
Packet number 91282 ending of length 82 at char number 9622337 failed checksum! Received 18 Calculated 91
Packet number 92906 ending of length 83 at char number 9793273 failed checksum! Received 36 Calculated 220
Packet number 94433 ending of length 75 at char number 9953935 failed checksum! Received 58 Calculated 68
Packet number 94460 ending of length 82 at char number 9956754 failed checksum! Received 142 Calculated 14
Packet number 102956 ending of length 74 at char number 10850686 failed checksum! Received 126 Calculated 147
Data stream statistics :

Packets and checksums :
115004 packets were found
114979 had good checksums
25 had incorrect checksums

General issues :
110456 leading characters were dropped
4683 false starts occurred
4681 double start bytes occurred
2 stray part packets occurred
172 chars lost from false starts 

Escaped byte profile :
36467 escape bytes were found
11232 escaped stop bytes were found
13244 escaped start bytes were found
11991 escaped escape bytes were found
0 escape pairs were mismatched

Report counts for all non-zero packet types
    Packet of type 0x0001 / 1 was found 2 times!
    Packet of type 0x0003 / 3 was found 2 times!
    Packet of type 0x0101 / 257 was found 36 times!
    Packet of type 0x0105 / 261 was found 31 times!
    Packet of type 0x0109 / 265 was found 2 times!
    Packet of type 0x0191 / 401 was found 49337 times!
    Packet of type 0xda5f / 55903 was found 2 times!
    Packet of type 0xf8e1 / 63713 was found 31 times!

Good packet length stats :
11735519 total sum of good packet bytes
102 average good packet length (with remainder of 7661)
FreeAir:bin fred$ 
This makes me wonder... are you checking checksums??? I see no messages saying that checksums are fucked, only that lengths are wrong which is EXTREMELY unlikely on a random basis like that, but as you know it can happen consistently if I have a bug :-) So, I reckon you're not checking them ;-) I could be wrong, if so, why does my decoder find bad packets and yours not? There were 21 buggy packets, according to you, and one resend at the start. Maybe the 3 other bad ones were streamed and ignored by you? I'm just speculating, but it all "seems" reasonable :-)

Notes:

When changing a value in 3d it is updated in 2d fine. The reverse is not true, but I noticed the button that says "reset display" after this, is that what it does? If so, why not automatic? If not what does it do?

When tuning in 3d (only practical thing due to no cell highlighting in 2d yet) the values get updated in the 2d, BUT, the colours are all wrong, and "LOL" :-) I have screenshots, I'll upload one here soon and if I don't bug me for it, I have them...

When you resize 3d it goes funky, maybe yellow/buffer issues from the graphics driver? Unsure, didn't get a screeny, but maybe I could if I resized a bunch and took lots of screenies.

I didn't do the save/restore, but I might go out and do that now once I record what I did find.

Tuning was mostly a success, we leaned out the and flattened the idle region and it became a LOT better, hopefully when its cold it's OK too. Will try that tomorrow and log it. What wasn't a success was loss of sync. I hope my debug vars for this run were usefully assigned as I'll need to fix that before I do much more. It's hard to know what AFR you've got when it's not firing the injectors or plugs half the time. I noted that it was output 3 and 4 that dropped most, this is a function of where it loses sync. 4 dropped the most, 3 the next most. I can use the angles in the code to work out when it was losing it and the numbers in the log (hopefully) to see how it was losing them.

I may need to finish the decoder to improve upon this.

Cranking is the same, I noticed that it fcuks up on and off while cranking causing it to not catch immediately. I'll have to solve that too.

i didn't even try, but will the fancy gauges work for me? It'd be really useful to highlight JUST the AFR/EGO while watching the spot on the 3d map (or 2d when you add highlighting) and a gauge is ONE way to do that. Perhaps there are others I don't know about?

Code: Select all

FreeAir:~ fred$ megatunix
Xlib:  extension "RANDR" missing on display "/tmp/launch-riW56S/org.x:0".
timeout, no packet found in queue for sequence 24 (18), locID 261
Re-issuing command!
Re-issued command sent, seq 46!

(megatunix:215): Gtk-WARNING **: Unable to find default local directory monitor type
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 81
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 81
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 82
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 75
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 74
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 81
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 70
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 81
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 72
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 81
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 82
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 71
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 82
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 81
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 78
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 82
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 82
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 83
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 0, Payload Length 781, raw pkt len 75
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 82
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 74
Packet fields don't make sense!
Can you treat a file as a serial device by chance? If so, you could rerun the bin file.

OK, Going outside to do a save and restore. And maybe a table export too.

Fred.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Fred's first MTX tuning session!

Post by Fred »

Yeah, looking at those lengths it's clear that you don't check checksums! Naughty Dave! :-)

Your packet type and sequence sorting and length checking etc should NOT be done until after the checksum is checked...

Time for a little refactoring :-)

Fred.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Fred's first MTX tuning session!

Post by Fred »

There is a little bug in my parser too that I just noticed. 115000 packets reported, and the sum of actual packets at the end is circa 50k so its incrementing the counter when it finds a start, instead of a stop. I'll have to fix that.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Fred's first MTX tuning session!

Post by Fred »

40 logs per second, and they are currently FAT compared to how big they need to be. OR, they could have more useful stuff in them. So good. I guess I'll have to fix up Gufi's OLV bin code to parse this as it'll have a mix of requested logs and streamed logs with different header contents. I'll add some debug on bad packets to OLV too. MegaSquirt can't touch this logging level with their current arch. Nice. I just need to sort out the time stamps and sequencing so that you can see the gaps between each record and it'll be GREAT.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Fred's first MTX tuning session!

Post by Fred »

ROFL:

Code: Select all

-copy-persistence-xml:

-compile-depend:

-do-compile:

-post-compile:

compile:

run:
     [java] Exception in thread "CVS Type Log Loading" java.lang.OutOfMemoryError: Java heap space
     [java] 	at java.util.Arrays.copyOf(Arrays.java:2734)
     [java] 	at java.util.ArrayList.ensureCapacity(ArrayList.java:167)
     [java] 	at java.util.ArrayList.add(ArrayList.java:351)
     [java] 	at java.util.regex.Pattern.split(Pattern.java:1003)
     [java] 	at java.lang.String.split(String.java:2292)
     [java] 	at java.lang.String.split(String.java:2334)
     [java] 	at Decoder.CSVTypeLog.decodeLog(CSVTypeLog.java:75)
     [java] 	at Decoder.BaseDecoder.run(BaseDecoder.java:44)
     [java] 	at java.lang.Thread.run(Thread.java:637)

DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Fred's first MTX tuning session!

Post by Fred »

Well, backup and restore work! Time for those screenies :

I think these funky colours are a bug:

http://stuff.fredcooke.com/mtx.2d.after ... olours.png
http://stuff.fredcooke.com/mtx.2d.after ... olours.png
http://stuff.fredcooke.com/mtx.2d.after ... olours.png
http://stuff.fredcooke.com/mtx.2d.after ... olours.png

And I think these are just cool to look at, and interesting:

Image

Image

Image

Image

Image

Image

Fred.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
dandruczyk
LQFP112 - Up with the play
Posts: 100
Joined: Sun Apr 06, 2008 6:30 pm

Re: Fred's first MTX tuning session!

Post by dandruczyk »

Fred wrote:This thread is to document the findings of a single half hour session driving around in the rain trying to sort the car out a bit better.

Firstly, thanks dave, MTX was a pleasure to use, mostly! :-)

X11 on the mac, freezing, not such a pleasure (3 times, had to reboot machine...).

Console afterwards had this in it, I'm going to try to find the binary logs now (on by default I hope, if not, [fizzle]... oh well, next time...) to run through my parser and compare with this output. I took a few logs, hopefully MTX writes out the "custom" ones in real time such that when the machine dies you get whatever you get. Yep, one input log of 12megs and 20 minutes duration ctime/mtime. Got that, will run it through my parser now and paste results.

Code: Select all

FreeAir:bin fred$ ./a.out /Users/fred/first.tuning.session/inputlog-2011-03-05_2135.bin 
The length of file /Users/fred/first.tuning.session/inputlog-2011-03-05_2135.bin is 12119661

Attempting to parse file...

Packet number 11425 ending of length 81 at char number 1217502 failed checksum! Received 169 Calculated 199
Packet number 13228 ending of length 81 at char number 1407476 failed checksum! Received 95 Calculated 43
Packet number 20748 ending of length 82 at char number 2199613 failed checksum! Received 233 Calculated 203
Packet number 22088 ending of length 75 at char number 2340590 failed checksum! Received 157 Calculated 228
Packet number 24396 ending of length 186 at char number 2583443 failed checksum! Received 188 Calculated 83
Packet number 28047 ending of length 74 at char number 2968051 failed checksum! Received 200 Calculated 47
Packet number 34458 ending of length 81 at char number 3641678 failed checksum! Received 121 Calculated 127
Packet number 40035 ending of length 187 at char number 4227664 failed checksum! Received 27 Calculated 221
Packet number 44545 ending of length 70 at char number 4702466 failed checksum! Received 220 Calculated 128
Packet number 46189 ending of length 81 at char number 4875753 failed checksum! Received 35 Calculated 155
Packet number 47486 ending of length 72 at char number 5012176 failed checksum! Received 62 Calculated 201
Packet number 49366 ending of length 81 at char number 5209840 failed checksum! Received 226 Calculated 198
Packet number 60276 ending of length 82 at char number 6358648 failed checksum! Received 33 Calculated 186
Packet number 66225 ending of length 181 at char number 6985892 failed checksum! Received 116 Calculated 156
Packet number 67089 ending of length 71 at char number 7076813 failed checksum! Received 68 Calculated 107
Packet number 68687 ending of length 82 at char number 7244854 failed checksum! Received 168 Calculated 41
Packet number 69890 ending of length 81 at char number 7371601 failed checksum! Received 102 Calculated 110
Packet number 83477 ending of length 78 at char number 8801037 failed checksum! Received 88 Calculated 73
Packet number 84425 ending of length 102 at char number 8900744 failed checksum! Received 79 Calculated 190
Packet number 85041 ending of length 82 at char number 8965547 failed checksum! Received 132 Calculated 111
Packet number 91282 ending of length 82 at char number 9622337 failed checksum! Received 18 Calculated 91
Packet number 92906 ending of length 83 at char number 9793273 failed checksum! Received 36 Calculated 220
Packet number 94433 ending of length 75 at char number 9953935 failed checksum! Received 58 Calculated 68
Packet number 94460 ending of length 82 at char number 9956754 failed checksum! Received 142 Calculated 14
Packet number 102956 ending of length 74 at char number 10850686 failed checksum! Received 126 Calculated 147
Data stream statistics :

Packets and checksums :
115004 packets were found
114979 had good checksums
25 had incorrect checksums

General issues :
110456 leading characters were dropped
4683 false starts occurred
4681 double start bytes occurred
2 stray part packets occurred
172 chars lost from false starts 

Escaped byte profile :
36467 escape bytes were found
11232 escaped stop bytes were found
13244 escaped start bytes were found
11991 escaped escape bytes were found
0 escape pairs were mismatched

Report counts for all non-zero packet types
    Packet of type 0x0001 / 1 was found 2 times!
    Packet of type 0x0003 / 3 was found 2 times!
    Packet of type 0x0101 / 257 was found 36 times!
    Packet of type 0x0105 / 261 was found 31 times!
    Packet of type 0x0109 / 265 was found 2 times!
    Packet of type 0x0191 / 401 was found 49337 times!
    Packet of type 0xda5f / 55903 was found 2 times!
    Packet of type 0xf8e1 / 63713 was found 31 times!

Good packet length stats :
11735519 total sum of good packet bytes
102 average good packet length (with remainder of 7661)
FreeAir:bin fred$ 
This makes me wonder... are you checking checksums??? I see no messages saying that checksums are [fizzle], only that lengths are wrong which is EXTREMELY unlikely on a random basis like that, but as you know it can happen consistently if I have a bug :-) So, I reckon you're not checking them ;-) I could be wrong, if so, why does my decoder find bad packets and yours not? There were 21 buggy packets, according to you, and one resend at the start. Maybe the 3 other bad ones were streamed and ignored by you? I'm just speculating, but it all "seems" reasonable :-)
See the source, yes I am and yes I'm not displaying the info on it

Code: Select all

 /* Check that the checksum matches */
                                if(checksum != lastChar)
                                {
                                        badChecksums++;
                                        /*printf("Packet number %u ending of length %u at char number %u failed checksum! Received %u Calculated %u\n", packets, currentPacketLength, processed, lastChar, checksum);*/
                                }
Fred wrote: Notes:

When changing a value in 3d it is updated in 2d fine. The reverse is not true, but I noticed the button that says "reset display" after this, is that what it does? If so, why not automatic? If not what does it do?
Yes I am aware of this, haven't found out why yet
Fred wrote: When tuning in 3d (only practical thing due to no cell highlighting in 2d yet) the values get updated in the 2d, BUT, the colours are all wrong, and "LOL" :-) I have screenshots, I'll upload one here soon and if I don't bug me for it, I have them...
Yes I know, but a get data should reset the colors
Fred wrote: When you resize 3d it goes funky, maybe yellow/buffer issues from the graphics driver? Unsure, didn't get a screeny, but maybe I could if I resized a bunch and took lots of screenies.
Click "reset display" a couple times and see if it fixes it.
I see this on some systems with shit opengl support (intel, virtualbox, etc), but never on nvidia
Fred wrote:
I didn't do the save/restore, but I might go out and do that now once I record what I did find.

Tuning was mostly a success, we leaned out the and flattened the idle region and it became a LOT better, hopefully when its cold it's OK too. Will try that tomorrow and log it. What wasn't a success was loss of sync. I hope my debug vars for this run were usefully assigned as I'll need to fix that before I do much more. It's hard to know what AFR you've got when it's not firing the injectors or plugs half the time. I noted that it was output 3 and 4 that dropped most, this is a function of where it loses sync. 4 dropped the most, 3 the next most. I can use the angles in the code to work out when it was losing it and the numbers in the log (hopefully) to see how it was losing them.

I may need to finish the decoder to improve upon this.

Cranking is the same, I noticed that it fcuks up on and off while cranking causing it to not catch immediately. I'll have to solve that too.

i didn't even try, but will the fancy gauges work for me? It'd be really useful to highlight JUST the AFR/EGO while watching the spot on the 3d map (or 2d when you add highlighting) and a gauge is ONE way to do that. Perhaps there are others I don't know about?
Gauges will only work if you create a custom dash (dashdesigner) and link it to the freeems data source names,. (within dashdesigner). the vetable display should be connected to them, if not the Gui/vetable-16x16x3-FreEMS.datamap.in file need to be adjusted to name the correct datasources.
Fred wrote:

Code: Select all

FreeAir:~ fred$ megatunix
Xlib:  extension "RANDR" missing on display "/tmp/launch-riW56S/org.x:0".
timeout, no packet found in queue for sequence 24 (18), locID 261
Re-issuing command!
Re-issued command sent, seq 46!

(megatunix:215): Gtk-WARNING **: Unable to find default local directory monitor type
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 81
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 81
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 82
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 75
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 74
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 81
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 70
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 81
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 72
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 81
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 82
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 71
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 82
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 81
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 78
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 82
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 82
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 83
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 0, Payload Length 781, raw pkt len 75
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 82
Packet fields don't make sense!
BAD PACKET!
payload length + header/payload EXCEEDS packet length, BUGGY PACKET!!
Payload ID: 401, Payload Length 96, raw pkt len 74
Packet fields don't make sense!
Can you treat a file as a serial device by chance? If so, you could rerun the bin file.

OK, Going outside to do a save and restore. And maybe a table export too.

Fred.
The BAD packet reports are from the packet decoder in mtx which does sanity checking, it's displaying the ones that had a bad checksum, it just displays it differently than you expected, i'll bet there's the same number ofthem there as reported by your tool. Mtx's packet decoder is a near carbon copy of your packet audit tool and its not structured in a way to cleanly disgard the packet at that time, so I let the packet get passed on where the decoder see's its fucked anyways. NO you can NOT use a file as a serial port, completely different semantics, no ioctls and lots of other stuff..
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Fred's first MTX tuning session!

Post by Fred »

Thanks for all the info!

The only thing I have to say is that it is fundamentally incorrect to do ANYTHING with the packet once the checksum fails. You should count it and drop it at that point. You certainly should NOT parse it further and report bogus errors as they are meaningless and worthless once the checskum has failed. I'd like to see "checksum failed" where it has, and wrong length where it is, but NEVER both, they are mutually exclusive.

Fred.
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
dandruczyk
LQFP112 - Up with the play
Posts: 100
Joined: Sun Apr 06, 2008 6:30 pm

Re: Fred's first MTX tuning session!

Post by dandruczyk »

Fred wrote:Thanks for all the info!

The only thing I have to say is that it is fundamentally incorrect to do ANYTHING with the packet once the checksum fails. You should count it and drop it at that point. You certainly should NOT parse it further and report bogus errors as they are meaningless and worthless once the checskum has failed. I'd like to see "checksum failed" where it has, and wrong length where it is, but NEVER both, they are mutually exclusive.

Fred.

Fixed. console output only for now..
User avatar
Fred
Moderator
Posts: 15431
Joined: Tue Jan 15, 2008 2:31 pm
Location: Home sweet home!
Contact:

Re: Fred's first MTX tuning session!

Post by Fred »

Awesome! :-)
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!
Post Reply