[PATCH 3/4] libblock: Change error status to fatal error

Chris Johns chrisj at rtems.org
Thu Apr 12 07:40:18 UTC 2012


On 12/04/12 5:27 PM, Sebastian Huber wrote:
>
> Not calling the init routine is the only why to trigger these fatal
> errors

You and I know this, then again we have both worked on the code.

> (except from system corruption like stack overflow etc.).

Understood.

>
> We already have special fatal errors for the different lock/unlock
> operations:
>
> #define RTEMS_BLKDEV_FATAL_BDBUF_SYNC_LOCK RTEMS_BLKDEV_FATAL_ERROR(11)
> #define RTEMS_BLKDEV_FATAL_BDBUF_SYNC_UNLOCK RTEMS_BLKDEV_FATAL_ERROR(12)
> #define RTEMS_BLKDEV_FATAL_BDBUF_CACHE_LOCK RTEMS_BLKDEV_FATAL_ERROR(13)
> #define RTEMS_BLKDEV_FATAL_BDBUF_CACHE_UNLOCK RTEMS_BLKDEV_FATAL_ERROR(14)
>

Can the status code in rtems_bdbuf_lock be tested for RTEMS_UNSATISFIED and

#define RTEMS_BLKDEV_FATAL_BDBUF_NO_INIT RTEMS_BLKDEV_FATAL_ERROR(xx)

be raised ?

Chris



More information about the devel mailing list