Pavel Pisa commented on a discussion on cpukit/dev/can/sja1000/sja1000.c: https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/1183#note_148190 > + "sja1000", > + irq_option, > + sja1000_interrupt, > + chip > + ); > + if ( sc != RTEMS_SUCCESSFUL ) { > + free( internal ); > + free( chip->qends_dev ); > + free( chip ); > + return NULL; > + } > + > + sc = rtems_task_create( > + rtems_build_name( 'C', 'A', 'N', 'S' ), > + worker_priority, > + RTEMS_MINIMUM_STACK_SIZE + 0x1000, Some stack usage statistic from i386_pc686 ``` SHLL [/] # stackuse STACK USAGE BY THREAD ID NAME LOW HIGH CURRENT AVAIL USED 0x09010001 IDLE 0x001b79d8 0x001b89bf 0x001b8918 4072 232 0x0a010002 RMON 0x001bb1f4 0x001bd1ef 0x001bcd2c 8188 1892 0x0a010003 SHLL 0x001bd208 0x001bf1ff 0x001be37c 8184 3812 0x0a010004 CANS 0x001bf28c 0x001c127f 0x001c10ac 8180 700 0x0a010005 CANS 0x001c12a0 0x001c328f 0x001c30bc 8176 692 0x0a010006 CANS 0x001c32b4 0x001c52af 0x001c50dc 8188 564 0x0a010007 CANS 0x001c52c8 0x001c72bf 0x001c70ec 8184 564 0x0a010008 CAND 0x001c72dc 0x001c92cf 0x001c91cc 8180 356 0x00000000 Interrupt Stack 0x001b69d0 0x001b79af 0x00000000 4064 1072 ``` It is 32-bit but register count is restricted so more local variables on stack. The first two `CANS` are SJA1000 workers actively used, the the other two `CANS` are only configured but not started. `CAND` is the background thread for postponed queues removal/free. So there is lot space but on the other hand, it is question how exceptions, task switching and interrupts occupy additional space on stack at different concurrent events. But main interrupt stack is separate so it should no be so much. -- View it on GitLab: https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/1183#note_148190 You're receiving this email because of your account on gitlab.rtems.org.
_______________________________________________ bugs mailing list [email protected] http://lists.rtems.org/mailman/listinfo/bugs
