linux-kernel - Re: USB: hub: Delete an error message for a failed memory allocation in usb_hub_clear_tt

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Thu, 7 Dec 2017 12:30:25 +0300
From:   Dan Carpenter <dan.carpenter@...cle.com>
To:     Geert Uytterhoeven <geert@...ux-m68k.org>
Cc:     Alan Stern <stern@...land.harvard.edu>,
        SF Markus Elfring <elfring@...rs.sourceforge.net>,
        USB list <linux-usb@...r.kernel.org>,
        Joe Perches <joe@...ches.com>,
        Daniel Drake <drake@...lessm.com>,
        Dmitry Fleytman <dmitry@...nix.com>,
        Eugene Korenevsky <ekorenevsky@...il.com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Günter Röck <linux@...ck-us.net>,
        Johan Hovold <johan@...nel.org>,
        Mathias Nyman <mathias.nyman@...ux.intel.com>,
        Peter Chen <peter.chen@....com>,
        LKML <linux-kernel@...r.kernel.org>,
        "kernel-janitors@...r.kernel.org" <kernel-janitors@...r.kernel.org>
Subject: Re: USB: hub: Delete an error message for a failed memory allocation
 in usb_hub_clear_tt_buffer()

On Thu, Dec 07, 2017 at 09:45:38AM +0100, Geert Uytterhoeven wrote:
> >
> > Small allocations never fail in the current kernel.
> 
> A few comments (this is in response to a patch from Markus, so there have
> to be lots of questions and uncertainties ;-)
> 1. In the current kernel. What about the future?

Right.  No one can predict.  And the small allocations don't fail rule
causes some problems.

> 2. If a small allocation cannot fail, what happens if the small memory slab
>    is exhausted? A new page must be allocated, which will trigger an OOM,
>    and some other part of the system will be killed and fail.

Right.


> 3. This driver uses GFP_ATOMIC, is that guaranteed to succeed? I think not.
> 

Right again.  I was missing the first email in the thread because of my
email filters so I didn't see this was atomic.

regards,
dan carpenter