S32DS under Linux flashing problem

cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

S32DS under Linux flashing problem

1,757 Views
przemyslaw_wiec
Contributor I

Hello All,

I have faced with S32 Design Studio (MPC) problem executing under Linux. It doesn't matter if it is version 1.2 or 2017R1. 

I try to flash MPC5748G via opensda interface executing command 'Flash from file' from S3DS level.. While this action executed under windows works fine under linux it is not. It stops at 90% and after some time I receive timeout information related to connection with gdb server ( which should be executed by S32DS i suppose ).

Doing the same under windows i got log information:

Connection from "127.0.0.1" via 127.0.0.1
Copyright 2017 P&E Microcomputer Systems,Inc.
Command Line :C:\NXP\S32DS_Power_v2017.R1\eclipse\plugins\com.pemicro.debug.gdbjtag.ppc_1.7.2.201709281658\win32\pegdbserver_power_console -device=MPC5748G -startserver -singlesession -serverport=7224 -gdbmiport=6224 -interface=OPENSDA -speed=5000 -port=Ě
 
CMD>RE
 
Initializing.
MPC574xC Device detected. 
Target has been RESET and is active.
CMD>CM C:\NXP\S32DS_Power_v2017.R1\eclipse\plugins\com.pemicro.debug.gdbjtag.ppc_1.7.2.201709281658\win32\gdi\P&E\nxp_mpc5748g_1x32x1520k_cflash.pcp
 
Initializing.
MPC574xC Device detected. 
Initialized.
 
;version 1.05, 06/26/2017, Copyright P&E Microcomputer Systems, www.pemicro.com [5748G_6080k]
 

(.... )

while under linux this info doesn't appear.

Could someone help me with this issue?

Best regards

Przemek

0 Kudos
Reply
6 Replies

1,471 Views
jiri_kral
NXP Employee
NXP Employee

Hi, 

it looks that you are using old version of PE Micro plugin - 1.7.2 . Current version is 1.7.5. Here is output from my console: 

pastedImage_1.png

Try to update your S32DS - Click on Help -> Install new Software. Click on Available Software Sites and make sure that PE Micro is enabled: 

pastedImage_2.png

Select All Available Sites in Work with combo box and continue with update. 

pastedImage_3.png

Hope it helps. 

Jiri 

0 Kudos
Reply

1,471 Views
przemyslaw_wiec
Contributor I

Thank You very much.

I have done above update procedure step by step with success, Unfortunately flashing behavior doesn't changed - still hanging. I attached screenshot i receive after some time ( where: Przekroczony czas oczekiwania na polaczenie = Connection timeout ). It looks everything is clear but I don't know why i get this while i have clear installation - there is no info related to additional steps needed for eg. gdb server configuration or something.

screenshot_2018-11-20_21-27-11.png

Below snapshot from console :

sreenshot_2018-11-20_21-37-27.png

And lsmod execution output ( device is detected ):

screenshot_2018-11-20_21-40-04.png

So there are any other ideas?

My OS is Slackware , kernel 4.19.4,  KDE5 but while s32ds is extended eclipse it should not be a case.

BR and thx for any help

P.W.

0 Kudos
Reply

1,471 Views
jiri_kral
NXP Employee
NXP Employee

Hi, 

can you publish full output from your debug console? 

Jiri

0 Kudos
Reply

1,471 Views
przemyslaw_wiec
Contributor I

I found reason but i'm not sure why it is like that,

I use DEVKIT-MPC5748G, which is equipped with 3 core MCU :  Z2_3 ,Z4_1,Z4_2.

Assuming I have 3 separate core application ( hello example ) - i can flash just when i have selected first core Z4_1 - Then I'm able select appropriate debug interface OpenSDA in flash windows ( Debug tab). When I try to do this for other core - flashing suck in one place (90%).

Similar situation is when i'm going to connect for debug - i can just debug core Z4_1, but while i try to debug other cores - debugger is going to hang on 90% as described above.

It seems S32DS can debug just first core - or there should be additional settings to perform debug on other cores.

I have the same situation is on WIN and on LINUX so it is common issue. I'll try dig into documentation but I anyone have idea what to do in that matter i'll be very appreciated.

BR.

p.w.

0 Kudos
Reply

1,471 Views
jiri_kral
NXP Employee
NXP Employee

Hi, 

the Z4_1 is boot core and is starting other cores. There is no way how to use this MCU without Z4_1 running (uploading .elf for that core). 

You can start debug with Z4_1 core - this config uploads also additional elf for other cores: 

pastedImage_1.png

Or  you can use launch group:

pastedImage_2.png

Jiri 

1,471 Views
przemyslaw_wiec
Contributor I

Jiri Thank You very much for your help,

regarding how it processor work i have some knowledge. So I know that core0 is a kind of master for other cores. For eg. hello example i have screen as You presented above so in this case I have expected that code for cores is flashed even it is done from level you show on the picture.

I didn't dig into code but I assumed that in case of presented example all cores have some job to do but when I put BP on core1 or core2 nothing happened. 

So then I have decided to test merged app ( i don't remember exact name of example ) but I this case i get same results - so when i put BP on core1 or core2 nothing happened.

I have to check launch groups You have presented maybe this is issue in my case. I'll check and will go back with feedback.

BR

P.W.

0 Kudos
Reply