[Starpu-devel] registering handles

[Hatem Ltaief <hatem.ltaief@kaust.edu.sa>]

Dear StarPU dev,

I am facing this error:

/home/hltaief/codes/starpu/install/lib/libstarpu-1.2.so.3(_starpu_select_src_node+0x241)[0x7ffff7254f91]

/home/hltaief/codes/starpu/install/lib/libstarpu-1.2.so.3(_starpu_create_request_to_fetch_data+0xd78)[0x7ffff72560c8]

/home/hltaief/codes/starpu/install/lib/libstarpu-1.2.so.3(_starpu_fetch_data_on_node+0x149)[0x7ffff7256559]

/home/hltaief/codes/starpu/install/lib/libstarpu-1.2.so.3(_starpu_fetch_task_input+0x166)[0x7ffff7256f76]

/home/hltaief/codes/starpu/install/lib/libstarpu-1.2.so.3(_starpu_cpu_driver_run_once+0x1fa)[0x7ffff7293d6a]

/home/hltaief/codes/starpu/install/lib/libstarpu-1.2.so.3(_starpu_cpu_worker+0x2d)[0x7ffff729439d]

/lib/x86_64-linux-gnu/libpthread.so.0(+0x76ba)[0x7fffe3f486ba]

/lib/x86_64-linux-gnu/libc.so.6(clone+0x6d)[0x7ffff4a7741d]

[starpu][_starpu_select_src_node][assert failure] The data for the handle 0xb1f6c10 is requested, but the handle does not have a valid value. Perhaps some initialization task is missing?

tb-rtm: datawizard/coherency.c:67: _starpu_select_src_node: Assertion `src_node_mask != 0' failed.

Thread 3 "CPU 0" received signal SIGABRT, Aborted.

[Switching to Thread 0x7fffd6eaa700 (LWP 32394)]

0x00007ffff49a5428 in raise () from /lib/x86_64-linux-gnu/libc.so.6

(gdb) bt

#0  0x00007ffff49a5428 in raise () from /lib/x86_64-linux-gnu/libc.so.6

#1  0x00007ffff49a702a in abort () from /lib/x86_64-linux-gnu/libc.so.6

#2  0x00007ffff499dbd7 in ?? () from /lib/x86_64-linux-gnu/libc.so.6

#3  0x00007ffff499dc82 in __assert_fail () from /lib/x86_64-linux-gnu/libc.so.6

#4  0x00007ffff7254fe1 in _starpu_select_src_node (handle=handle@entry=0xb1f6c10, destination=destination@entry=0) at datawizard/coherency.c:67

#5  0x00007ffff72560c8 in _starpu_create_request_to_fetch_data (handle=handle@entry=0xb1f6c10, dst_replicate=dst_replicate@entry=0xb1f6cf0, mode=mode@entry=STARPU_R, is_prefetch=is_prefetch@entry=0, 

    async=async@entry=0, callback_func=callback_func@entry=0x0, callback_arg=0x0) at datawizard/coherency.c:568

#6  0x00007ffff7256559 in _starpu_fetch_data_on_node (handle=handle@entry=0xb1f6c10, node=node@entry=0, dst_replicate=0xb1f6cf0, mode=mode@entry=STARPU_R, detached=detached@entry=0, 

    is_prefetch=is_prefetch@entry=0, async=0, callback_func=0x0, callback_arg=0x0) at datawizard/coherency.c:786

#7  0x00007ffff7256f76 in fetch_data (mode=STARPU_R, replicate=<optimized out>, node=0, handle=0xb1f6c10) at datawizard/coherency.c:814

#8  _starpu_fetch_task_input (j=j@entry=0xb1f7ef0) at datawizard/coherency.c:1004

#9  0x00007ffff7293d6a in execute_job_on_cpu (perf_arch=<optimized out>, rank=<optimized out>, cpu_args=0x7ffff75a7490 <_starpu_config+1520>, worker_task=0xb1f7cb0, j=0xb1f7ef0)

    at drivers/cpu/driver_cpu.c:74

#10 _starpu_cpu_driver_run_once (cpu_worker=cpu_worker@entry=0x7ffff75a7490 <_starpu_config+1520>) at drivers/cpu/driver_cpu.c:307

#11 0x00007ffff729439d in _starpu_cpu_worker (arg=0x7ffff75a7490 <_starpu_config+1520>) at drivers/cpu/driver_cpu.c:368

#12 0x00007fffe3f486ba in start_thread () from /lib/x86_64-linux-gnu/libpthread.so.0

#13 0x00007ffff4a7741d in clone () from /lib/x86_64-linux-gnu/libc.so.6

I am calling my MORSE task like this:

    for (int nt=0;nt<timestep;nt++) {

        for (int nd=0;nd<ndomains;nd++) {                                                            

           printf("\nnt %d nd %d %d %d\n", nt, nd, Unew->mb, Unew->nb);                                        

           MORSE_TASK_iow( Unew, nd, 0, nx, ny, nz, nt );                                            

        }                                                                                            

    } 

and the task is defined as follows:

void MORSE_TASK_iow(MORSE_desc_t *U, int i, int j, int nx, int ny, int nz, int timestep)              

{                                                                                                     

  struct starpu_codelet *codelet=&cl_iow;                                                             

                                                                                                      

  MORSE_BEGIN_ACCESS_DECLARATION;                                                                     

  MORSE_ACCESS_R(U, i, j);                                                                            

  MORSE_END_ACCESS_DECLARATION;                                                                       

                                                                                                      

/*                                                                                                    

*/                                                                                                    

  starpu_insert_task(                                                                                 

      starpu_mpi_codelet(codelet),                                                                    

      STARPU_R    , RTBLKADDR(U,float,i,j),                                                           

      STARPU_VALUE, &nx         ,sizeof(int),                                                         

      STARPU_VALUE, &ny         ,sizeof(int),                                                         

      STARPU_VALUE, &ny         ,sizeof(int),                                                         

      STARPU_VALUE, &timestep   ,sizeof(int),                                                         

      0);                                                                                             

}

Any clue on what is wrong? This may be a chameleon related issue.

Thanks,

H

---

This message and its contents including attachments are intended solely for the original recipient. If you are not the intended recipient or have received this message in error, please notify me immediately and delete
 this message from your computer system. Any unauthorized use or distribution is prohibited. Please consider the environment before printing this email.