Re: [RFC v1 07/11] net/mlx4: Cache irq_desc->affinity instead of irq_desc

From: Thomas Gleixner <tglx@linutronix.de>
To: Amir Vadai <amirv@mellanox.com>
Cc: Jiang Liu <jiang.liu@linux.intel.com>,
	Bjorn Helgaas <bhelgaas@google.com>,
	Benjamin Herrenschmidt <benh@kernel.crashing.org>,
	Ingo Molnar <mingo@redhat.com>, "H. Peter Anvin" <hpa@zytor.com>,
	"Rafael J. Wysocki" <rjw@rjwysocki.net>,
	Randy Dunlap <rdunlap@infradead.org>,
	Yinghai Lu <yinghai@kernel.org>, Borislav Petkov <bp@alien8.de>,
	Ido Shamay <idos@mellanox.com>,
	"David S. Miller" <davem@davemloft.net>,
	Or Gerlitz <ogerlitz@mellanox.com>,
	Eric Dumazet <edumazet@google.com>,
	Hadar Hen Zion <hadarh@mellanox.com>,
	Eran Ben Elisha <eranbe@mellanox.com>,
	Joe Perches <joe@perches.com>,
	Saeed Mahameed <saeedm@mellanox.com>,
	Matan Barak <matanb@mellanox.com>,
	Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>,
	Tony Luck <tony.luck@intel.com>,
	x86@kernel.org,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	linux-pci@vger.kernel.org, linux-acpi@vger.kernel.org
Subject: Re: [RFC v1 07/11] net/mlx4: Cache irq_desc->affinity instead of irq_desc
Date: Mon, 4 May 2015 17:10:26 +0200 (CEST)	[thread overview]
Message-ID: <alpine.DEB.2.11.1505041708190.4225@nanos> (raw)
In-Reply-To: <CAPcc5Ph4pk+4aL-PduvF67YU-A+vn2bfx6v3mr-tbY58zs7sZw@mail.gmail.com>

On Mon, 4 May 2015, Amir Vadai wrote:

> On Mon, May 4, 2015 at 6:15 AM, Jiang Liu <jiang.liu@linux.intel.com> wrote:
> > The field 'affinity' in irq_desc won't change once the irq_desc data
> > structure is created. So cache irq_desc->affinity instead of irq_desc.
> > This also helps to hide struct irq_desc from device drivers.
> 
> Hi Jiang,
> 
> I might not understand the new changes irq core, but up until now
> affinity was changed when the user changed it through
> /proc/irq/<IRQ>/smp_affinity.
> This code is monitoring the affinity from the napi_poll context to
> detect affinity changes, and prevent napi from keep running on the
> wrong CPU.
> Therefore, the affinity can't be cached at the beginning. Please
> revert this caching.

Sigh. All of this is completely wrong. Polling in the internals of
irq_desc is just crap.

irq_set_affinity_notifier() has been added for exactly this purpose.

Please get rid of your irq_desc abuse.

Thanks,

	tglx