Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Wed, 28 Aug 2019 09:45:25 +0800
From:   Ming Lei <ming.lei@redhat.com>
To:     Bart Van Assche <bvanassche@acm.org>
Cc:     Thomas Gleixner <tglx@linutronix.de>, Jens Axboe <axboe@fb.com>,
        Hannes Reinecke <hare@suse.com>,
        Sagi Grimberg <sagi@grimberg.me>, linux-scsi@vger.kernel.org,
        Peter Zijlstra <peterz@infradead.org>,
        Long Li <longli@microsoft.com>,
        John Garry <john.garry@huawei.com>,
        linux-kernel@vger.kernel.org, linux-nvme@lists.infradead.org,
        Keith Busch <keith.busch@intel.com>,
        Ingo Molnar <mingo@redhat.com>, Christoph Hellwig <hch@lst.de>
Subject: Re: [PATCH 3/4] nvme: pci: pass IRQF_RESCURE_THREAD to
 request_threaded_irq
Message-ID: <20190828014524.GA8090@ming.t460p>
References: <20190827085344.30799-1-ming.lei@redhat.com>
 <20190827085344.30799-4-ming.lei@redhat.com>
 <7cdb9dbb-46e5-b66a-ddf1-c7ecceb28d7a@acm.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <7cdb9dbb-46e5-b66a-ddf1-c7ecceb28d7a@acm.org>
User-Agent: Mutt/1.11.3 (2019-02-01)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Tue, Aug 27, 2019 at 08:10:42AM -0700, Bart Van Assche wrote:
> On 8/27/19 1:53 AM, Ming Lei wrote:
> > If one vector is spread on several CPUs, usually the interrupt is only
> > handled on one of these CPUs.
> 
> Is that perhaps a limitation of x86 interrupt handling hardware? See also
> the description of physical and logical destination mode of the local APIC
> in the Intel documentation.
> 
> Does that limitation also apply to other platforms than x86?

Please see the following excellent explanation from Thomas.

	https://lkml.org/lkml/2018/4/4/734

Especially the following words:

	So at some point we ripped out the multi target support on X86 and moved
	everything to single target delivery mode.
	
	Other architectures never supported multi target delivery either due to
	hardware restrictions or for similar reasons why X86 dropped it. There
	might be a few architectures which support it, but I have no overview at
	the moment.


thanks,
Ming