Re: [PATCH] zram: fix possible use after free in zcomp_create()

From: Luis Henriques
Date: Mon Sep 07 2015 - 10:11:44 EST


On Mon, Sep 07, 2015 at 10:33:32PM +0900, Sergey Senozhatsky wrote:
> On (09/07/15 13:53), Luis Henriques wrote:
> > > On (09/07/15 11:33), Luis Henriques wrote:
> > > > zcomp_create() verifies the success of zcomp_strm_{multi,siggle}_create()
> > > > through comp->stream, which can potentially be pointing to memory that was
> > > > freed if these functions returned an error.
> > > >
> > >
> > > good catch.
> > >
> > > we probably better start checking the zcomp_strm_multi_create()/
> > > zcomp_strm_single_create() status in zcomp_create(); that's much
> > > more obvious and that's why we return it in the first place.
> > >
> > > what do you think?
> > >
> >
> > Yep, that's probably a better solution.
> >
> > Acked-by: Luis Henriques <luis.henriques@xxxxxxxxxxxxx>
> >
>
> Oh, thanks. I don't mind if you will re-submit it; I just did minor
> changes to our fix. Or I can handle it. I'm OK with either way.

Ok, I'll be sending v2 in a minute.

>Btw, did you hit that problem or you reviewed the code?
>

I've found this while looking at the code (I wouldn't really call it a
"code review"... :-))

Cheers,
--
Luís
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/