From: arne@vajhoej.dk   
      
   On 12/4/2025 2:49 PM, Simon Clubley wrote:   
   > 1) What is the reason for using the CMA interface instead of the pthread   
   > interface ? If there is no specific reason, perhaps rewriting it as a   
   > pthreads application might give some clues.   
      
   I already have working pthreads code.   
      
   I wanted to do the same with cma.   
      
   Same reason that people climb MtEverest.   
      
   > 2) I assume cma_thread_create() and cma_thread_join() both return status   
   > codes ? (pthread does). Try capturing them and displaying them.   
      
   No.   
      
   void   
   cma_thread_create (   
             cma_t_thread   *new_thread,   
             cma_t_attr   *attr,   
             cma_t_start_routine   start_routine,   
             cma_t_address   arg);   
      
   void   
   cma_thread_join (   
             cma_t_thread   *thread,   
             cma_t_exit_status   *exit_status,   
             cma_t_address   *result);   
      
   > 3) Have you tried my suggestion to put 5 second pauses in the code so   
   > you have a chance to try and find out which resource is being consumed ?   
      
   Yes (not timed pause but press return to continue).   
      
   Neither SHOW PROC/QUO or SHOW PROC/ACC show anything critical.   
      
   Page file quota remaining goes down, so it is leaking memory.   
      
   But there are 2 million left when it crashes.   
      
   I wonder if I am running out of thread stack space.   
      
   Anyway no matter what is causing the crash - it should not leak.   
      
   When the thread is terminated and joined, then anything allocated   
   at create should be released.   
      
   Arne   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)