On 24/2/21 6:13 pm, Sebastian Huber wrote: > On 23/02/2021 23:41, Chris Johns wrote: > >> On 24/2/21 9:26 am, dufa...@hda.com wrote: >>> I re-read Joel's mail and I agree, the priority should be left ridiculously >>> low (as it is now) or maybe set in the middle (but why bother?). >>> >>> I was thinking about matching classic RTEMS behavior. I don't think it >>> matters in POSIX. >> The SYSINIT and constructors run on the first created thread before it is >> enteretd so it makes me wonder if we test initialisation with both types of >> init >> task? > The SYSINIT doesn't run in a task context. It runs in whatever context is set > up > by the BSP low level initialization.
Ah OK and thanks. I had not realised it worked this way but it makes sense. This limits what SYSINT offers over constructors but SYSINIT is a much better means of specifying and performing these initialisation operations. There is no sequencing in constructors. > Global construction is tested for all initialization task variants. Yes so it can use any RTEMS resource. >> Using the same priority would avoid any complication. > Maybe we should add configuration options for the POSIX initialization thread. +1 Chris _______________________________________________ devel mailing list devel@rtems.org http://lists.rtems.org/mailman/listinfo/devel
