stuck in OSIF_TimeDelay()

キャンセル
次の結果を表示 
表示  限定  | 次の代わりに検索 
もしかして: 

stuck in OSIF_TimeDelay()

ソリューションへジャンプ
3,711件の閲覧回数
thomasbao
Contributor II

Hello, I am using MPC5748G. I have a problem about OSIF_TimeDelay(), it will stuck in it, and the debug window looks like this:

thomasbao_0-1632903827565.png

 

After osif component is added, what else needs to be done to use OSIF_TimeDelay()?

Thanks.

0 件の賞賛
返信
1 解決策
3,593件の閲覧回数
lukaszadrapa
NXP TechSupport
NXP TechSupport

Hi,

After further investigation, I found out that it is known issue in S32DS debugger. I tested your project using Lauterbach debugger and it was working in this case.

It's not fixed yet but the workaround is to not use freeze mode - FRZ bit in PIT_MCR should be 0.

It works as expected in standalone mode without a debugger.

Sorry for inconvenience.

Regards,

Lukas

元の投稿で解決策を見る

0 件の賞賛
返信
9 返答(返信)
2,423件の閲覧回数
Francescovgg1
Contributor I

Hi, 

Is there an update on this topic?

I'm using MCP5748G with SDK RTM 3.0.3 and the problem is still present. When I'm in debug, I have to manually reset the FRZ bit via the embsys registers and the FRZ bit gets set back to 1, next time round.

Thank you for your help

0 件の賞賛
返信
2,924件の閲覧回数
davidanderle
Contributor II

I have the same issue and I use PIT as my system timer. Therefore, when the OSIF reconfigures the PIT peripheral, the FRZ bit is set and while my debugger is attached, my 1ms task is not triggered.

OSIF-PIT dependency is not listed in the referenced components. there are 2 ways to stop this from happening: Use the STM module for the system timer or patch the osif driver manually (comment out FRZ bit set)

2,859件の閲覧回数
Bobkat
Contributor II

Hi David,

Apologies for the late reply.

Ended up debugging over CAN. Not ideal, but seemed to get the job done.

Will change over to the STM module for timing when I next work on the core code.

Many thanks

Rob

0 件の賞賛
返信
3,615件の閲覧回数
thomasbao
Contributor II

Hi,

    I used MPC5748G-GW-RDB, the pin I measured is shown in the red box below, and there are no other originals attached to J11.

thomasbao_0-1634115895983.png

 

0 件の賞賛
返信
3,594件の閲覧回数
lukaszadrapa
NXP TechSupport
NXP TechSupport

Hi,

After further investigation, I found out that it is known issue in S32DS debugger. I tested your project using Lauterbach debugger and it was working in this case.

It's not fixed yet but the workaround is to not use freeze mode - FRZ bit in PIT_MCR should be 0.

It works as expected in standalone mode without a debugger.

Sorry for inconvenience.

Regards,

Lukas

0 件の賞賛
返信
3,438件の閲覧回数
Bobkat
Contributor II

Hi Lukas,

Encountering this problem with S32ds, MPC5748G and OSIF.

With regard to the workaround, we're using the SDK and can only reset the bit via the embsys registers and the FRZ bit gets set back to 1, next time round.

The application doesn't recognise PIT.MCR.B.FRZ = 0; it looks like 'PIT' is redefined by the SDK.

Any ideas as to how to set the FRZ bit to 0 permanently?

Additionally, is there a schedule for releasing an update to S32ds that will fix this problem?

Best regards

 

Rob

 

0 件の賞賛
返信
3,703件の閲覧回数
lukaszadrapa
NXP TechSupport
NXP TechSupport

Hi,

it works as expected on my side without additional configuration. Do you have some simple project to reproduce the behavior?

Regards,

Lukas

0 件の賞賛
返信
3,642件の閲覧回数
thomasbao
Contributor II

Hi,

    Here is a simple project to flip the level of PA0 every 100ms, but the level of PA0 remains 0V after entering the while loop, even when observed with an oscilloscope.

    Could you give me some advice? Thank you.

0 件の賞賛
返信
3,627件の閲覧回数
lukaszadrapa
NXP TechSupport
NXP TechSupport

Hi,

it works as expected on my side. I used DEVKIT-MPC5748G and I can see that USR_LED2 connected to PA0 is toggled. Do you measure correct pin?

Regards,

Lukas

0 件の賞賛
返信