[openib-general] kernel oops

[Hal Rosenstock]

On Tue, 2005-08-30 at 19:50, Sean Hefty wrote:
> Hal Rosenstock wrote:
> >>Can we just remove this field and 
> >>use the sgid to locate the correct device structure in the kernel, or 
> >>fail if it cannot be located?
> > 
> > That seems like a good idea.
> 
> Quickly skimming through the code I couldn't easily locate where AT maintained a 
> device list, or how it retrieved the device pointer.

AT tracks IPoIB netdevices rather than IB devices but one can get at the
IB device through the ipoib_dev_priv structure which is available
through the netdevice.

> > Won't AT still be needed under the new CM abstraction for IB ? I guess
> > the answer is unclear. It still seems to me that it should be fixed
> > until there is something else to take its place. Do you concur ?
> 
> Had the fix been easy (for me to figure out how to make anyway) I would have 
> submitted a patch.  Something like AT is likely to be needed, but it's not clear 
> how close the final version will be to what's there now.  If we can at least 
> validate the device pointer, it may be good enough to continue using for the 
> time being.

I think it is possible to validate the device pointer in the route
rather than change the API. I'll work on a patch for this.

-- Hal