[PATCH] Fix PR 21337 (v4): segfault when re-reading symbols with remote debugging.

[Doug Gilmore]

On 06/27/2017 10:28 AM, Doug Gilmore wrote:
> On 06/25/2017 04:24 AM, Simon Marchi wrote:
>> Hi Doug,
>>
>> Good thing you pinged, I had completely missed it, sorry for the wait.
>>
>> +gdb_test "source $gdbfile" ".*source-command-completed.*" \
>> +    "source $testfile.gdb"
>> +# Sometimes the failure only occurs on the second invocation.
>> +gdb_test "source $gdbfile" ".*source-command-completed.*" \
>> +    "source $testfile.gdb"
>>
>> Tests should have unique names, so that when one fails, you know easily which one.  So, perhaps "source $testfile.gdb 1" and "source $testfile.gdb 2"?
>>
>> The patch is ok with this fixed.
>>
>> Thanks,
>>
>> Simon
> Hi Simon and Maciej,
> 
> Simon: I updated the tests per your request.  When I asked Maciej to
> commit the patch for me, he noticed that the comment extended over the
> soft 72 column limit of, so I reformatted it.  Also I corrected a typo
> and format issues in the changelog entries that he noticed.
> 
> Thanks,
> 
> Doug
Also I forgot to mention per Maciej's request, that my work is covered
by a copyright assignment to FSF.

Doug