Debug SAS DLL for testing

[poppo]

...., which now leads me to think you may not be running the latest dll!

Not sure if any paths are hard coded, but Bunker's screenshot shows it's not intalled in the default install folder. Notice the dll is in 'test app' folder.

[AndyP]

Well according to microsoft you are not allowed to redistribute debug libraries... so I will put the dlls up on my site here http://users.tpg.com.au/adslcmdn/SASEGMTest/debugdll/msvcr100d.dll just in case I need to use them to debug some stuff on a customer site.

I am pretty sure that you can just put it in the same directory as the dll, if not is c:\windows\system32

[poppo]

I am pretty sure that you can just put it in the same directory as the dll,

That works. Program now runs and talks to the machine (but old version already did for me). Could Bunker's issue be related to the dll folder noted above?

[AndyP]

I am pretty sure that you can just put it in the same directory as the dll,

That works. Program now runs and talks to the machine (but old version already did for me). Could Bunker's issue be related to the dll folder noted above?

most likely. New dll's need to go in the same folder as SASEGMTest.exe

[poppo]

dbmon working too. Here is a small capture. What's up with line 4032 - 'Chirp from Poll Address: 0'.  Address 0?  Normal? I get those groups of chirps frequently, but only that one (for the short time I let it run) had address 0. All the other groups were all address 1.


4032:   Enter: CSerialPrinterPort::WriteCommBlock
4032:     Leave: ReadCommBlock
4032:     Bytes Read 0
4032:     Empty buffer!
4032:   Leave: CSerialPrinterPort::WriteCommBlock
4032:     Waiting on comm/shutdown events
4032:   Write General Poll
4032:   Enter: CSerialPrinterPort::WriteCommBlock
4032:   Leave: CSerialPrinterPort::WriteCommBlock
4032:     Comm signalled
4032:     Enter: ReadCommBlock
4032:       Parity Error
4032:       IO Pending
4032:     Leave: ReadCommBlock
4032:     Bytes Read 25
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 1
4032:     Chirp from Poll Address: 0
4032:     Waiting on comm/shutdown events
4032:     Comm signalled
4032:     Enter: ReadCommBlock
4032:       IO Pending
4032:   Retreive General Response
4032:   Enter: RetrieveResponse
4032:   Leave: RetrieveResponse
4032:   Write Sequence Poll
4032:   Enter: CSerialPrinterPort::WriteCommBlock
4032:   Leave: CSerialPrinterPort::WriteCommBlock
4032:   Write General Poll
4032:   Enter: CSerialPrinterPort::WriteCommBlock
4032:   Leave: CSerialPrinterPort::WriteCommBlock
4032:     Leave: ReadCommBlock
4032:     Bytes Read 1

[AndyP]

No, thats one of the problems with SAS and PC's, its not actually a chirp, its a no activity response from a general poll. The problem is, you can see that 25 bytes were in the buffer, and when you read them out there is no way to tell if the parity error is for one or all. If you could, all the chirps wouls show parity error (because the wakeup bit is set) and the response would show no parity error.

I will get the timing stuff sorted soon, and that should fix most of those little issues.

[stayouttadabunker]

I am pretty sure that you can just put it in the same directory as the dll,

That works. Program now runs and talks to the machine (but old version already did for me). Could Bunker's issue be related to the dll folder noted above?

most likely. New dll's need to go in the same folder as SASEGMTest.exe

Okay,
I just stuck everything into the SASEGMTest.exe folder...
I'll go try it again.


ADD>>> No go Houston...Here's a screenshot...
Everything's in the same folder...>>>

[poppo]

ADD>>> No go Houston...Here's a screenshot...
Everything's in the same folder...>>>

Ok, are you sure that is the new dll? If it is you should have been getting the errors we were getting without the msvcr100d.dll. And I don't see that in there. If you go to details view of the folder, the dll file size should be about 279k vice 56k of the old one.

[stayouttadabunker]

I think so...it was larger in size if I recall...279kb

[poppo]

I think so...it was larger in size if I recall...279kb

Never mind, I see it's highlighted and the size is shown. But are you running the SASEGMTest.exe from that directory too? Or using a shortcut pointing to a different one?

Maybe the solution is 

 

[stayouttadabunker]

yup...same size as yours...lol
Below that is the "Properties" folder>>>

[stayouttadabunker]

I think so...it was larger in size if I recall...279kb

Never mind, I see it's highlighted and the size is shown. But are you running the SASEGMTest.exe from that directory too? Or using a shortcut pointing to a different one?

Maybe the solution is 

 

Nope, I run it from the "Start" button on the left>>>

[poppo]

I don't know if you should have another copy in the properties directory???  And I put the msvcr100d.dll in the SAGEGMTest folder.

Try double clicking the SASEGMtest.exe right from within the folder.

[stayouttadabunker]

Yeah...I did but this pops up instead>>

[poppo]

Yeah...I did but this pops up instead>>

That's good since that is what we were getting. Now move the msvcr100d.dll into the SAGEGMTest folder and try again (double clicking from within the folder).

[stayouttadabunker]

okay,
I re-arranged my folders to look exactly like yours.>>>

[poppo]

okay,
I re-arranged my folders to look exactly like yours.>>>

Ok, now double click the program from within the folder and.....<drum roll>....

[stayouttadabunker]

<Smash!> drums busts...lol

[stayouttadabunker]

good night Poppo...
you tried...
I gotta  
I just verified that the serial cable is good because the Sim still works fine...

 

[poppo]

Did you double click the SASEGMtest.exe or the msvcr100d.dll?  Should be clicking the SASEGMtest.exe

Oh well, tomorrow is another day. Wait...it is tomorrow. 

[poppo]

Ok, after getting a good night's sleep , the only other thing I noticed is that it looks like you are using all of the files from the original program and not the first updated one. AndyP will have to answer this, but maybe this dll will not work right with those other files. I am going by the file sizes and time/dates that your screenshot shows vs the ones on mine.

Once that is sorted out and if nothing else, the dbmon program should now work. The output of dbmon is pretty fast, so after you verify it is outputting something, you may want to close it and then run it like this for about 10 seconds:
dbmon >c:\dbmon.log
That will direct the output to the dbmon.log file which you can then attach for AndyP to look at. Note: you won't see any output on the screen when run tihis way, it's all going to the log file.

[AndyP]

Ok, after getting a good night's sleep , the only other thing I noticed is that it looks like you are using all of the files from the original program and not the first updated one. AndyP will have to answer this, but maybe this dll will not work right with those other files. I am going by the file sizes and time/dates that your screenshot shows vs the ones on mine.

Once that is sorted out and if nothing else, the dbmon program should now work. The output of dbmon is pretty fast, so after you verify it is outputting something, you may want to close it and then run it like this for about 10 seconds:
dbmon >c:\dbmon.log
That will direct the output to the dbmon.log file which you can then attach for AndyP to look at. Note: you won't see any output on the screen when run tihis way, it's all going to the log file.

Best way to work out version is by the build date and size. I am well overdue for sleep (midnight here) so will take a look in the morning.

Directing the dbmon output to file is the best way. All I need is the first couple off polls and maybe a request meters 10-15 then quit the app. The poll rate is pretty quick, so wont take long to make a huge file.

[stayouttadabunker]

lol...Good day...
I've tried double clicking all kinds of ways to start the App.
I've even run it straight from the "run" commands and
still cannot get past the [01:] displayed.
After a fresh night's sleep, I'm ready to attack this again but
from a different computer....however with an S+ this time.
Last night it was my S2000.

I just sent an email to AndyP showing him screenshots of the working Sim from last nite.
There seems to be some problem with they way it's
downloading the TestApp during installation perhaps.
I'm going to try it again with this computer and run the DBMON and the other one.
I'm also going to re-arrange the file folders of the APP on this computer similar to yours.

[poppo]

I'm also going to re-arrange the file folders of the APP on this computer similar to yours.

The main problem I see after your last screen-shot is the wrong versions of the support files. Not sure if that matters or not, but since the second version was not just a DLL change like this update, I would assume it does.

Sometimes having multiple copies of different versions will cause problems if the wrong files are being loaded. That is sort of why you were not getting the error message like you 'should' have after updating the dll.

I agree that you should try it on a S+. The meters button won't work for some SP chips (it calls for meters that don't exist on older SPs), but you should get door open/close messages.

[stayouttadabunker]

I agree totally with you.
I'm making up another harness like the first one
I made for the S+ Interface board  (J2 to the serial PC port)
as I still have a not-burned-out working S+ interface board one here...lol