alpha c++ test failures

alpha c++ test failures

[Richard Henderson]

While none of the C tests I have examined so far have been particularly
serious, the C++ check does not complete because it appears that both
the rtti and exception handling tests frequently produce executables
that run indefinitely.

This, then, is a request for a change to the test suite that will limit
the cpu time use by any one test, perhaps by doing a "ulimit -t 10"
before starting the test. 

Not knowing much about the way the test harness is structured, this is a
stab in the dark, but it seems that any such change also provide for a
per-test option to override the default time (which should be short) for
cases that are expected to take some extended amount of time to complete.
But this is definitely not the case for the vast majority of the tests.


r~