• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 99 subscribers
  • Views 242 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

MessageQ creation required why?

RCReddy
Genius 3575 points

Hi  All,

           Using CCSV5, MCSDK [bios_mcsdk_02_00_05_17_setupwin32],i started with a project of sending a message from CORE0 to CORE1. Aim of the project is achieved [MessageQ_Put and MessageQ_Get is successfull] and in the meanwhile process, i found few things which i am not able to understand

I am creating a queue at CORE1 and opening it in CORE0 [to do MessageQ_put]. So queue is local to CORE1 and remote to CORE0.

My code is working only if i have created a local queue at CORE0 also [Though i am not using this CORE0 local queue]. 

=================================================

messageQ = MessageQ_create(localqueue,NULL);

status1 = MessageQ_open(remotequeue,&remoteQueueId);

HeapBufMP_Params_init(&HeapBufparams);

MyMsg = (MessageQMsg *)MessageQ_alloc(HEAPID,sizeof(MyMsg));

=================================================

if you look at above steps [CORE0 Code], MessageQ_open is on remote queue whereas i am also creating a localqueue [if i comment the local MessageQ_create line, i see the simulation crashing].My question is "IS IT REQUIRED/NECESSARY THAT FOR A SHARED REGION heapmembuf based IPC mechanism to have both messageQ's created at both cores [if i have only one queue created at core1 and opened by core0, will it not work..is it mandatory that i have to define Q's at both cores]".

Thanks

R C Reddy

  • 0
    TI__Guru* 96960 points

    Hi,

    MessageQ does not require a queue to be create on every core. When you comment out Core0's creation of the Message Queue, do you also remove the MessageQ_open and associated logic from Core1?

    Todd

    [edited on 4/27...added the not]

  • 0 in reply to ToddMullanix
    Genius 3575 points

    Hi Todd,

                  I have only one way of communication implemented [that is core0 opens a remotequeue and puts message thats all] [core1 has created that remotequeue] and i have not implemented the reverse logic [that is core0 creates a queue which will be opened by core1 ][this is the case which you are talking about and i have not implemented it ].

    The other thing what i noticed is instead of MessageQ_create on CORE0, if i put Task_Sleep(1), the code works without $c$exit. so between IPC_attach and MessageQ_open i already have one Task_sleep [as part of Do While loop for ipc_attach] and further i introduced one more Task_sleep(1) before MessageQ_open [earlier there was MessageQ_create which was introducing some sort of delay now that i have replaced it by Task_sleep(1)], code is not crashing in simulation..

    Todd,

             Any idea why is this behaviour

    Thanks

    RC Reddy

  • 0 in reply to RCReddy
    TI__Guru* 96960 points

    RC,

    Are you looping on the MessageQ_open until it succeeds? If core1 has not created the queue yet, the MessageQ_open will fail, so you need to loop until it succeeds.

    Todd