linux-kernel - Re: [PATCH 5/5] drivers/pci: make most of the PCI

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]

Message-ID: <20160208155141.GT23093@windriver.com>
Date:	Mon, 8 Feb 2016 10:51:41 -0500
From:	Paul Gortmaker <paul.gortmaker@...driver.com>
To:	Arnd Bergmann <arnd@...db.de>
CC:	<linux-kernel@...r.kernel.org>,
	Bjorn Helgaas <bhelgaas@...gle.com>,
	Ley Foon Tan <lftan@...era.com>,
	Geert Uytterhoeven <geert+renesas@...der.be>,
	Stanimir Varbanov <svarbanov@...sol.com>,
	Thierry Reding <thierry.reding@...il.com>,
	<linux-pci@...r.kernel.org>, <linux-arm-msm@...r.kernel.org>
Subject: Re: [PATCH 5/5] drivers/pci: make most of the PCI_DW drivers modular

[Re: [PATCH 5/5] drivers/pci: make most of the PCI_DW drivers modular] On 08/02/2016 (Mon 11:00) Arnd Bergmann wrote:

> On Sunday 07 February 2016 19:00:44 Paul Gortmaker wrote:
> > We've exported the symbols that we know these specific drivers
> > will need as tristate, so now we can make the conversion from
> > bool to tristate w/o concern for build regressions.
> > 
> > Cc: Bjorn Helgaas <bhelgaas@...gle.com>
> > Cc: Ley Foon Tan <lftan@...era.com>
> > Cc: Geert Uytterhoeven <geert+renesas@...der.be>
> > Cc: Stanimir Varbanov <svarbanov@...sol.com>
> > Cc: Thierry Reding <thierry.reding@...il.com>
> > Cc: Arnd Bergmann <arnd@...db.de>
> > Cc: linux-pci@...r.kernel.org
> > Cc: linux-arm-msm@...r.kernel.org
> > Signed-off-by: Paul Gortmaker <paul.gortmaker@...driver.com>
> 
> Looks good, but I'm not sure what happens on unload. Did you check?

Per the 0/N I've not got the capability to do run time testing for any
of this stuff.  And from previous discussions, ISTR that people didn't
want to see a blanket block-unload policy implemented.

> Until this is tested, maybe it's best to prevent module unloading?

If that is what people want, I can add that.

P.
-- 

> 
> 	Arnd