Sebastian Huber commented on a discussion on bsps/sparc/leon3/start/eirq.c: https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/1514#note_160097 > > #ifdef LEON3_IRQAMP_IRQMAP > /* Mapping from bus interrupt lines to IRQ(A)MP interrupt lines */ > -rtems_vector_number LEON3_IrqCtrl_Mapping[ BSP_INTERRUPT_VECTOR_MAX_MAP + 1 > ]; > +rtems_vector_number LEON3_IrqCtrl_Mapping[ BSP_INTERRUPT_VECTOR_COUNT ]; > + > +/* > + * A map field is eight bits wide. A field value above the last interrupt > + * line of the IRQ(A)MP selects no interrupt line. Line 0 is no interrupt. > + */ > +static rtems_vector_number leon3_irqmap_line( uint32_t field ) I removed the check. The BSP takes the field modulo BSP_INTERRUPT_VECTOR_COUNT, which compiles to a mask. Even a faulty register value then gives a valid dispatch table index and a defined bit shift. A comment states that such a value is a hardware fault. rtems/tools/rtems-sis!13 also stores a written field modulo 32 now. -- View it on GitLab: https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/1514#note_160097 You're receiving this email because of your account on gitlab.rtems.org. Unsubscribe from this thread: https://gitlab.rtems.org/-/namespace/49/sent_notifications/5-8frp48radtjwtwa5j0nazp34z-1d/unsubscribe | Manage all notifications: https://gitlab.rtems.org/-/profile/notifications | Help: https://gitlab.rtems.org/help
_______________________________________________ bugs mailing list [email protected] http://lists.rtems.org/mailman/listinfo/bugs
