Re: [PATCH 00/16 v3] Linux extended-remote fork and exec events

[Pedro Alves <palves@redhat.com>]

On 10/31/2014 11:28 PM, Don Breazeal wrote:
> 
>  - gdb.threads/thread-execl.exp gives a couple of failures related to
>    scheduler locking.  As with the previous item, after spending some
>    time on this I concluded that pursuing it further now would be
>    feature-creep, and that this should be tracked with a bug report.

Do you have more details on this?

Looking at the exec race you mentioned, I thought that thread-execl.exp should
expose it,  given that the point of the test is exactly a thread other than
the main thread execing.  But then I stumbled on the fact that running it with
your series on top of currently mainline often crashes gdb:

$ make check RUNTESTFLAGS="--target_board=native-extended-gdbserver thread-execl.exp"
...
Running /home/pedro/gdb/mygit/src/gdb/testsuite/gdb.threads/thread-execl.exp ...
ERROR: Process no longer exists

                === gdb Summary ===

# of expected passes            9
# of unresolved testcases       1

Odd that this doesn't trigger with native testing.

Thanks,
Pedro Alves