Avoid duplicate thread recycling.

[answer9030]

At the end of the test, the traffic thread has been reclaimed. If there is an exception in the control connection, it will cause the thread to be reclaimed repeatedly. Use sp->done to avoid repeated thread recycling.

PLEASE NOTE the following text from the iperf3 license. Submitting a
pull request to the iperf3 repository constitutes "[making]
Enhancements available...publicly":

You are under no obligation whatsoever to provide any bug fixes, patches, or
upgrades to the features, functionality or performance of the source code
("Enhancements") to anyone; however, if you choose to make your Enhancements
available either publicly, or directly to Lawrence Berkeley National
Laboratory, without imposing a separate written license agreement for such
Enhancements, then you hereby grant the following license: a non-exclusive,
royalty-free perpetual license to install, use, modify, prepare derivative
works, incorporate into other computer software, distribute, and sublicense
such enhancements or derivative works thereof, in binary and source code form.

The complete iperf3 license is available in the LICENSE file in the
top directory of the iperf3 source tree.

• Version of iperf3 (or development branch, such as master or
  3.1-STABLE) to which this pull request applies:

• Issues fixed (if any): segfault when multi thread repeated recycling in client #1760

• Brief description of code changes (suitable for use as a commit message):

[bmah888]

Thanks for the PR! I think the change makes sense...the only way that sp->done gets set to 1 would be if we're going to terminate the corresponding thread.

I'd probably just check for sp->done rather than a check for a specific value (sp->done == 1), this would be consistent with the other checks against sp->done.

There are two other places in iperf_client_api.c and one in iperf_server_api.c that do a similar iteration over all (or most) of the streams and try to pthread_cancel() their corresponding threads. Do you think that similar checks for sp->done would be appropriate in those cases as well?

[answer9030]

I think there is no need to perform similar checks on sp ->done in other places, as these places will not trigger duplicate thread recycling.

[bmah888]

The original scenario for #1760 is a little difficult for me to reproduce exactly, but thanks for the detailed explanation of what triggers the segfault. I settled for tested this by more-or-less forcing the error handling on the client to run. I replaced this code in iperf_client_api.c (function iperf_run_client):

		if (iperf_set_send_state(test, TEST_END) != 0)
                    goto cleanup_and_fail;

with:

iperf_set_test_state(test, TEST_END);
//		if (iperf_set_send_state(test, TEST_END) != 0)
                    goto cleanup_and_fail;

I was able to reproduce the original segfault and as well as verify that the client exits sanely.

THanks again for the PR!