The problem is in find_fhe. It seem that unused entries in the
cache have NULL dentry entries. This would cause find_fhe to return
an fhe that contained a NULL dentry and the next test would OOPs
when it executed
dentry = fhe->dentry;
inode = dentry->d_inode;
Allen
P.S. I mangled nfsbug.c to do these tests. I will try to make it pretty
enough to post ;-) soon.
>>>Bill Hawes said:
> G. Allen Morris III wrote:
>
> > Here is a newer patch for kernel.
> >
> > This has three changes:
> >
> > 3) An OOPS was caused if the dcookie in the nfs file handle was
> > NULL.
>
> Hi Allen,
>
> Under what conditions are you seeing a NULL dcookie in the filehandle? This
> shouldn't happen, and if an external client sent such a mangled fh it should
> fail the validation test.
>
> Do you have an example that triggers this problem?
>
---------------------------------
G. Allen Morris III
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/