Re: [Bug 218] Probable mmap() bug that breaks cpp-3.2

[Steve Lord <lord@xxxxxxx>]

On Tue, 2003-02-25 at 15:51, Chris Wedgwood wrote:
> On Tue, Feb 25, 2003 at 03:45:14PM -0600, Steve Lord wrote:
> 
> > Nope, the problem was this filesystem was last running a kernel from
> > the time period where that bug was present. So there were a bunch of
> > files in this state. Running mapcheck cleaned up the filesystems,
> > and a new kernel cleaned up the bug.
> 
> I'm confused.... i read it as, 'new kernel, will deadlock, mapcheck
> fixes this'.

Nope, old kernel created bad files, confused cpp in gcc 3.2, new kernel
does not generate these files. mapcheck fixes up old files/

> 
> So I assumed I could store data past EOF and still confuse a new
> kernel.  I was wrong I take it?

Yep.

Steve

-- 

Steve Lord                                      voice: +1-651-683-3511
Principal Engineer, Filesystem Software         email: lord@xxxxxxx