[Bug ada/38989] How much stack space should c380004 take?

["joel dot sherrill at oarcorp dot com" <gcc-bugzilla at gcc dot gnu dot org>]

------- Comment #2 from joel dot sherrill at oarcorp dot com  2009-01-27 20:10 -------
Subject: Re:  How much stack space should c380004 take?

laurent at guerby dot net wrote:
> ------- Comment #1 from laurent at guerby dot net  2009-01-27 19:31 -------
> This test has arrays of complicated Ada entities so I'm not surprised by stack
> usage. Could you report at what task size it passes?
>
>
>   
Passes with the Ada Init task having a stack of 1029K.
Fails when it is 1028K.  That is over a megabyte of stack.

The rest of the ACATS tests run with 100K init task stack
and a default Ada task stack size of 20K.  [That might be
40K if the run-time doubles the number still.]  Either way
still WAY less than 1 megabyte.

I can increase the stack size for the ACATS RTEMS initialization
to get by this. But it would be good to know if there is a lower
limit on the target which should be able to pass the ACATS.

Can't the test mechanics pass an argument to a single test case?
That might be an option -- increase the stack size just for this.

What do you think? 

--joel


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38989