trafficserver-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From James Peach <jpe...@apache.org>
Subject Re: git commit: [TS-3112] - Add null pointer check for contp to prevent core dump after handleEvent(TS_FETCH_EVENT_EXT_HEAD_DONE)
Date Fri, 03 Oct 2014 15:33:07 GMT
On Oct 3, 2014, at 8:28 AM, James Peach <jpeach@apache.org> wrote:

> On Oct 3, 2014, at 6:29 AM, sudheerv@apache.org wrote:
> 
>> Repository: trafficserver
>> Updated Branches:
>> refs/heads/master 33f651c90 -> d1b3dc66b
>> 
>> 
>> [TS-3112] - Add null pointer check for contp to prevent core dump after handleEvent(TS_FETCH_EVENT_EXT_HEAD_DONE)
>> 
>> 
>> Project: http://git-wip-us.apache.org/repos/asf/trafficserver/repo
>> Commit: http://git-wip-us.apache.org/repos/asf/trafficserver/commit/d1b3dc66
>> Tree: http://git-wip-us.apache.org/repos/asf/trafficserver/tree/d1b3dc66
>> Diff: http://git-wip-us.apache.org/repos/asf/trafficserver/diff/d1b3dc66
>> 
>> Branch: refs/heads/master
>> Commit: d1b3dc66b5725879949350890ab014cf235cae64
>> Parents: 33f651c
>> Author: Sudheer Vinukonda <sudheerv@yahoo-inc.com>
>> Authored: Fri Oct 3 13:29:03 2014 +0000
>> Committer: Sudheer Vinukonda <sudheerv@yahoo-inc.com>
>> Committed: Fri Oct 3 13:29:03 2014 +0000
>> 
>> ----------------------------------------------------------------------
>> proxy/FetchSM.cc | 3 +++
>> 1 file changed, 3 insertions(+)
>> ----------------------------------------------------------------------
>> 
>> 
>> http://git-wip-us.apache.org/repos/asf/trafficserver/blob/d1b3dc66/proxy/FetchSM.cc
>> ----------------------------------------------------------------------
>> diff --git a/proxy/FetchSM.cc b/proxy/FetchSM.cc
>> index d7b187a..4a79db4 100644
>> --- a/proxy/FetchSM.cc
>> +++ b/proxy/FetchSM.cc
>> @@ -249,6 +249,9 @@ FetchSM::InvokePluginExt(int fetch_event)
>>    has_sent_header = true;
>>  }
>> 
>> +  if (!contp)
>> +    goto out;
>> +
> 
> There's a check for contp being NULL just 10 lines above here ... how can it become NULL
now?

Also, InvokePluginExt weirdly mixes sending plugin events and de-chunking ... if there's no
continuation, don't you still need to de-chunk? Any why is there even an InvokePlugin and
InkvokePluginExt?

J
Mime
View raw message