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  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:	Mon, 7 Jul 2008 14:29:23 -0500
From:	Jack Steiner <steiner@....com>
To:	Arjan van de Ven <arjan@...radead.org>
Cc:	Hugh Dickins <hugh@...itas.com>,
	Christoph Hellwig <hch@...radead.org>,
	Nick Piggin <nickpiggin@...oo.com.au>, cl@...ux-foundation.org,
	akpm@...l.org, linux-kernel@...r.kernel.org, mingo@...e.hu,
	tglx@...utronix.de, holt@....com, andrea@...ranet.com,
	linux-mm@...ck.org
Subject: Re: [patch 12/13] GRU Driver V3 -  export is_uv_system(), zap_page_range() & follow_page()

On Mon, Jul 07, 2008 at 11:58:44AM -0700, Arjan van de Ven wrote:
> On Mon, 7 Jul 2008 17:29:54 +0100 (BST)
> Hugh Dickins <hugh@...itas.com> wrote:
> 
> > On Mon, 7 Jul 2008, Jack Steiner wrote:
> > > > > +EXPORT_SYMBOL_GPL(follow_page);
> > > > 
> > > > NACK.
> > > > 
> > > > These should never be called by a driver and suggest you need to
> > > > rething your VM integration in this driver.
> > > 
> > > Can you provide some additional details on the type of kernel API
> > > that could be exported to provide a pte lookup in atomic context?
> > 
> > I don't see EXPORT_SYMBOL_GPL(follow_page) as objectionable myself:
> > it rather seems rather to complement EXPORT_SYMBOL(vm_insert_page)
> > and EXPORT_SYMBOL(vmalloc_to_page); though I'd agree that it's
> > sufficiently sensitive to need that _GPL on it.
> > 
> > ...
> > 
> > > Currently, the driver calls follow_page() in interrupt context.
> > 
> > However, that's a problem, isn't it, given the pte_offset_map_lock
> > in follow_page?  To avoid the possibility of deadlock, wouldn't we
> > have to change all the page table locking to irq-disabling variants?
> > Which I think we'd have reason to prefer not to do.
> > 
> > Maybe study the assumptions Nick is making in his arch/x86/mm/gup.c
> > in mm, and do something similar in your GRU driver (falling back to
> > the slow method when anything's not quite right).  It's not nice to
> > have such code out in a driver, but GRU is going to be exceptional,
> > and it may be better to have it out there than pretence of generality
> > in the core mm exporting it.
> > 
> 
> I wonder if GRU even should be a module; it sounds rather like pretty
> core functionality and if it's this invasive to the VM it probably
> should be a real part of the VM instead

The GRU is not actually very invasive into the VM. It will use the
new MMU-notifier callbacks. Aside from the need to translate
virt->physical & zap ptes belonging to the GRU, it works fine as a module.
No other core changes are needed.

An additional advantage in keeping it as a module is that I expect it
to under a number of changes as the hardware matures. It is easier to
update the GRU if it is a module.

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

Powered by blists - more mailing lists