threading issues

[Daniel Eischen]

On Mon, 11 Oct 2004, Matt Hamilton wrote:

>
> All,
>    I am still trying to work out why pthread support on AMD64 appears
> broken.  (See message on -current 3/10/2004 "Re: 5.3-BETA5 AMD64 Python
> problems").
>
> I was looking in CVS, and saw a test suite in /usr/src/libpthread/test.
>   In running Make I get:
>
> ...compile stuff...
> Test static library:
> --------------------------------------------------------------------------
> Test                                      c_user c_system c_total     chng
>   passed/FAILED                            h_user h_system h_total   % chng
> --------------------------------------------------------------------------
> hello_d                                     0.00     0.00    0.00
>   passed
> --------------------------------------------------------------------------
> hello_s                                     0.00     0.01    0.01
>   passed
> --------------------------------------------------------------------------
> join_leak_d                                 0.04     0.00    0.04
>   *** FAILED ***

It doesn't really fail, the test just doesn't take into account the fact
that the library caches threads and KSEs.

> --------------------------------------------------------------------------
> mutex_d                                 ^C
>
> it just hangs on mutex_d and I had to manually abort it.  Does this work
> on i386 systems?  This is an AMD Opteron (HP DL145) running BETA7 as
> compiled from -current at the weekend.

It doesn't hang.  The test just takes a long time to run.  Run it manually
and see for yourself.

-- 
Dan Eischen