netdev - Re: [PATCH 1/7] 8139cp: Improve accuracy of cp_interrupt() return, to survive IRQ storms

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, 24 Sep 2015 00:18:10 +0100
From:	David Woodhouse <dwmw2@...radead.org>
To:	Francois Romieu <romieu@...zoreil.com>
Cc:	David Miller <davem@...emloft.net>, netdev@...r.kernel.org
Subject: Re: [PATCH 1/7] 8139cp: Improve accuracy of cp_interrupt() return,
 to survive IRQ storms

On Thu, 2015-09-24 at 00:44 +0200, Francois Romieu wrote:
> David Woodhouse <dwmw2@...radead.org> :
> [...]
> > But sure, for now I'll drop this from the series and I can try to
> > convince you separately.
> 
> You may as well change the IRQ storm avoidance logic so that it does not
> require conformant driver code if you do not want to waste more time :o)

That's not entirely trivial — at a certain level we *do* require that
drivers don't lie to the core IRQ logic.

I suppose even if a rogue driver is returning IRQ_HANDLED in an IRQ
storm, when it's *not* actually doing anything or making any progress,
perhaps there's a way we could detect that we're always going straight
back into the next interrupt handler as *soon* as we exit the previous
one. Perhaps there's merit in exploring that option.

I'm not sure that gives you a valid excuse for not wanting to fix the
driver though :)

I might start by "clarifying" the documentation on the IRQ handler's
return value...

-- 
David Woodhouse                            Open Source Technology Centre
David.Woodhouse@...el.com                              Intel Corporation

Download attachment "smime.p7s" of type "application/x-pkcs7-signature" (5691 bytes)