Kinsey Moore started a new discussion on 
bsps/riscv/esp32/include/c6/chip_definitions.h: 
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/1186#note_148270

 > + * SUCH DAMAGE.
 > + */
 > +
 > +#ifndef LIBBSP_ESP32_C6_CHIP_DEFINITIONS_H
 > +#define LIBBSP_ESP32_C6_CHIP_DEFINITIONS_H
 > +
 > +#ifndef ASM
 > +
 > +#include <bsp.h>
 > +
 > +#endif /* ASM */
 > +
 > +/*
 > + * References: 
 > https://documentation.espressif.com/esp32-c6_technical_reference_manual_en.pdf#intmtrx
 > + * The interrupt matrix supports 77 interrupts. The count of 78 below 
 > includes 77 peripheral interrupts plus
 > + * invalid interrupt 1?

It looks like this may vary on the C6. On the C3, interrupt 0 is reserved for 
exceptions. The C6 supports CLINT which occupies CPU interrupts 0,3,4,7 of 
which only 3 and 7 are important for M-mode operation. The implementation here 
will be more complicated than the C3 because the RTEMS-facing interrupt 
management will need to take these into account whereas on the C3 there were 
*only* external interrupts and no CLINT. This will likely affect the publicly 
advertised interrupt values below.

-- 
View it on GitLab: 
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/1186#note_148270
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

Reply via email to