genmcf548x patches

Joel Sherrill joel.sherrill at OARcorp.com
Fri Apr 20 11:15:54 UTC 2012


My previous email was wondering if it was covering up mismanagement of enable/disable fpu.

I don't have this board but there are only a handful of tests which would need to be run to check this. Classic hello, POSIX hello, sp19 (I think) and there might be one other test with float in the sptests. The state of the enable bits would need to be checked in fp and not fp tasks in the hellos since they don't explicitly use fpu and thus won't fault if wrong.

I think the fp save needs to enable/disable the fpu around its work and the restore needs to enable it and leave it on

Sebastian Huber <sebastian.huber at embedded-brains.de> wrote:

>For me it is completely unclear why we need a special case for initialization 
>tasks.  I think this special case handling shadows only the real problem.  What 
>happens if all initialization tasks are non-FP and you create and start a FP 
>task later?
>
>-- 
>Sebastian Huber, embedded brains GmbH
>
>Address : Obere Lagerstr. 30, D-82178 Puchheim, Germany
>Phone   : +49 89 18 90 80 79-6
>Fax     : +49 89 18 90 80 79-9
>E-Mail  : sebastian.huber at embedded-brains.de
>PGP     : Public key available on request.
>
>Diese Nachricht ist keine geschäftliche Mitteilung im Sinne des EHUG.
>_______________________________________________
>rtems-devel mailing list
>rtems-devel at rtems.org
>http://www.rtems.org/mailman/listinfo/rtems-devel




More information about the devel mailing list