archive mirror
 help / color / mirror / Atom feed
From: Sebastian Andrzej Siewior <>
To: Keith Busch <>
Subject: Re: [RFC PATCH 1/3] nvme/pci: Poll the cq in the primary irq handler
Date: Tue, 3 Dec 2019 11:09:30 +0100	[thread overview]
Message-ID: <> (raw)
In-Reply-To: <>

On 2019-12-03 07:22:04 [+0900], Keith Busch wrote:
> The nvme threaded interrupt handler reduces CPU time spent in hard irq
> context, but waking it increases latency for low queue depth workloads.
> Poll the completion queue once from the primary handler and wake the
> thread only if more completions remain after. Since there is a window
> of time where the threaded and primary handlers may run simultaneously,
> add a new nvmeq flag so that the two can synchronize which owns processing
> the queue.

It depends on what you mean by "run simultaneously" but it sounds like
this does not happen.

The primary handler disables the interrupt source and returns
IRQ_WAKE_THREAD. From now on, the primary handler won't fire (unless it
is a shared handler and someone else gets an interrupt).

The thread receives a wakeup and will be show on the CPU.
Once the threaded handler completes its work it enables the interrupt
source as the last thing it does. This may lead to an interrupt before
the threaded handler returns to the caller.
The primary handler may return IRQ_WAKE_THREAD. In this case the
threaded handler will be invoked again.

If it wouldn't then this could happen:

  primary handler                     threaded handler

      if (test_and_set_bit(NVMEQ_THREAD, &nvmeq->flags))
      ret = nvme_irq(irq, data);
      if (!nvme_cqe_pending(nvmeq))
      if (to_pci_dev(nvmeq->dev->dev)->msix_enabled)
         __pci_msix_desc_mask_irq(irq_get_msi_desc(irq), 1);
      return IRQ_WAKE_THREAD;
				       if (to_pci_dev(nvmeq->dev->dev)->msix_enabled)
				         __pci_msix_desc_mask_irq(irq_get_msi_desc(irq), 0);
      if (test_and_set_bit(NVMEQ_THREAD, &nvmeq->flags))
        return IRQ_NONE;

	                               clear_bit(NVMEQ_THREAD, &nvmeq->flags);
                                       return IRQ_HANDLED

> Signed-off-by: Keith Busch <>


linux-nvme mailing list

  parent reply	other threads:[~2019-12-03 10:09 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-12-02 22:22 [RFC PATCH 0/3] nvme threaded interrupt improvements Keith Busch
2019-12-02 22:22 ` [RFC PATCH 1/3] nvme/pci: Poll the cq in the primary irq handler Keith Busch
2019-12-03  7:50   ` Christoph Hellwig
2019-12-03  9:39     ` Sebastian Andrzej Siewior
2019-12-03 11:50     ` Keith Busch
2019-12-03 10:09   ` Sebastian Andrzej Siewior [this message]
2019-12-03 11:16     ` Keith Busch
2019-12-04 10:21       ` Ming Lei
2019-12-02 22:22 ` [RFC PATCH 2/3] nvme/pci: Remove use_threaded_interrupts parameter Keith Busch
2019-12-02 22:22 ` [RFC PATCH 3/3] nvme/pci: Poll for new completions in irq thread Keith Busch
2019-12-03 10:17   ` Sebastian Andrzej Siewior

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \ \ \ \ \ \ \ \

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).