[omniORB] Re: Purify UMR in 4.0.1

Duncan Grisby duncan at grisby.org
Fri Jul 18 12:19:50 BST 2003


On Friday 11 July, Andrew Pickin wrote:

> The following UMR occur wih gcc 2.95 and 3.3.0 builds of OmniOrb-4.0.1
> in the event of a failed connection.

What do you mean by failed connection?  The code in question has
nothing to do with connections -- it's the thing that finds the list
of the machine's network interfaces.

> Can someone please confirm that these are indeed omni faults?

No, it's not a fault. The uninitialised memory buffer in question is
filled in by an ioctl call. It's quite common for those parts of an OS
to not bother initialising all the bytes.

Are you actually encountering a problem?  If so, what is it?

Cheers,

Duncan.

-- 
 -- Duncan Grisby         --
  -- duncan at grisby.org     --
   -- http://www.grisby.org --



More information about the omniORB-list mailing list