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] [day] [month] [year] [list]
Message-ID: <20161007081136.1eb8fae9@vento.lan>
Date:   Fri, 7 Oct 2016 08:11:36 -0300
From:   Mauro Carvalho Chehab <mchehab@...pensource.com>
To:     Jiri Kosina <jikos@...nel.org>
Cc:     Jörg Otte <jrg.otte@...il.com>,
        Johannes Stezenbach <js@...uxtv.org>,
        Patrick Boettcher <patrick.boettcher@...teo.de>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Andy Lutomirski <luto@...nel.org>,
        Michael Krufky <mkrufky@...uxtv.org>,
        Mauro Carvalho Chehab <mchehab@...nel.org>,
        Linux Media Mailing List <linux-media@...r.kernel.org>
Subject: Re: Problem with VMAP_STACK=y

Em Fri, 7 Oct 2016 09:52:56 +0200 (CEST)
Jiri Kosina <jikos@...nel.org> escreveu:

> On Thu, 6 Oct 2016, Mauro Carvalho Chehab wrote:
> 
> > I can't see any other obvious error on the conversion. You could try to 
> > enable debug options at DVB core/dvb-usb and/or add some printk's to the 
> > driver and see what's happening.
> 
> Mauro, also please don't forget that there are many more places in 
> drivers/media that still perform DMA on stack, and so have to be fixed for 
> 4.9 (as VMAP_STACK makes that to be immediately visible problem even on 
> x86_64, which it wasn't the case before).

Yes, I'm aware of that. I'm doing the conversion of drivers under dvb-usb,
at:
	https://git.linuxtv.org/mchehab/experimental.git/log/?h=media_dmastack_fixes

I'll be sending the patches to the ML after ready.

I'll then take a look on other USB drivers that use the stack. I guess
the non-USB media drivers are safe from this issue.

Thanks,
Mauro

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ