Kernel memleak in jffs2_erase_block() (fs/jffs2/erase.c:78)

[Joakim Tjernlund]

On Mon, 2007-11-12 at 13:35 +0100, Jörn Engel wrote:
> On Mon, 12 November 2007 13:30:49 +0100, Joakim Tjernlund wrote:
> > 
> > Nope, instr is freed in jffs2_erase_callback() in this case.
> 
> Ick!  So this will double-free if an erase method calls the callback
> _and_ returns an error.  Given the well-documented interfaces we have, I
> wouldn't be surprised to find such a driver.
> 
> Oh well, just ignore my patch then.
> 
> Jörn

Ouch, I should read all my mail before posting replies :)

 Jocke