Re: [classlib/luni] check for invalid socket before I/O operations (HARMONY-6131)

[Oliver Deakin]

+1

Checked on Linux - I don't see any negative effects, and it fixes the 
RegistryTest crash and ConnectionTest EOFException failures for me.

Regards,
Oliver

Sean Qiu wrote:
+1

2009/3/30 Tim Ellison <t.p.elli...@gmail.com>:
  
I would like to propose that this is fixed for M9, since it is part of
the remaining blocker for declaring the milestone.

Any committers support applying this patch?

Thanks,
Tim

Tim Ellison (JIRA) wrote:
    
Tim Ellison reassigned HARMONY-6131:
------------------------------------

    Assignee: Tim Ellison

      
[classlib/luni] check for invalid socket before I/O operations
--------------------------------------------------------------

                Key: HARMONY-6131
                URL: https://issues.apache.org/jira/browse/HARMONY-6131
            Project: Harmony
         Issue Type: Bug
         Components: Classlib
           Reporter: Stanley Gambarin
           Assignee: Tim Ellison
        Attachments: OSNetworkSystem.c.patch


new socket I/O code does not check before read/write operations for invalid 
sockets (it used to).
This may cause a crash if we are trying to write to a closed/reset socket.
        



  

--
Oliver Deakin
Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number 741598. 
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU