From: Jacob Keller <jacob.e.keller@intel.com>
To: Jakub Kicinski <kuba@kernel.org>, Thomas Gleixner <tglx@linutronix.de>
Cc: Marcelo Tosatti <mtosatti@redhat.com>,
Nitesh Narayan Lal <nitesh@redhat.com>,
Peter Zijlstra <peterz@infradead.org>,
helgaas@kernel.org, linux-kernel@vger.kernel.org,
netdev@vger.kernel.org, linux-pci@vger.kernel.org,
intel-wired-lan@lists.osuosl.org, frederic@kernel.org,
sassmann@redhat.com, jesse.brandeburg@intel.com,
lihong.yang@intel.com, jeffrey.t.kirsher@intel.com,
jlelli@redhat.com, hch@infradead.org, bhelgaas@google.com,
mike.marciniszyn@intel.com, dennis.dalessandro@intel.com,
thomas.lendacky@amd.com, jiri@nvidia.com, mingo@redhat.com,
juri.lelli@redhat.com, vincent.guittot@linaro.org,
lgoncalv@redhat.com
Subject: Re: [PATCH v4 4/4] PCI: Limit pci_alloc_irq_vectors() to housekeeping CPUs
Date: Mon, 26 Oct 2020 15:52:46 -0700 [thread overview]
Message-ID: <63c3484d-327e-5f37-7860-3af277c26711@intel.com> (raw)
In-Reply-To: <20201026151306.4af991a5@kicinski-fedora-PC1C0HJN.hsd1.ca.comcast.net>
On 10/26/2020 3:13 PM, Jakub Kicinski wrote:
> On Mon, 26 Oct 2020 22:50:45 +0100 Thomas Gleixner wrote:
>> On Mon, Oct 26 2020 at 14:11, Jacob Keller wrote:
>>> On 10/26/2020 1:11 PM, Thomas Gleixner wrote:
>>>> On Mon, Oct 26 2020 at 12:21, Jacob Keller wrote:
>>>>> Are there drivers which use more than one interrupt per queue? I know
>>>>> drivers have multiple management interrupts.. and I guess some drivers
>>>>> do combined 1 interrupt per pair of Tx/Rx.. It's also plausible to to
>>>>> have multiple queues for one interrupt .. I'm not sure how a single
>>>>> queue with multiple interrupts would work though.
>>>>
>>>> For block there is always one interrupt per queue. Some Network drivers
>>>> seem to have seperate RX and TX interrupts per queue.
>>> That's true when thinking of Tx and Rx as a single queue. Another way to
>>> think about it is "one rx queue" and "one tx queue" each with their own
>>> interrupt...
>>>
>>> Even if there are devices which force there to be exactly queue pairs,
>>> you could still think of them as separate entities?
>>
>> Interesting thought.
>>
>> But as Jakub explained networking queues are fundamentally different
>> from block queues on the RX side. For block the request issued on queue
>> X will raise the complete interrupt on queue X.
>>
>> For networking the TX side will raise the TX interrupt on the queue on
>> which the packet was queued obviously or should I say hopefully. :)
>>
>> But incoming packets will be directed to some receive queue based on a
>> hash or whatever crystallball logic the firmware decided to implement.
>>
>> Which makes this not really suitable for the managed interrupt and
>> spreading approach which is used by block-mq. Hrm...
>>
>> But I still think that for curing that isolation stuff we want at least
>> some information from the driver. Alternative solution would be to grant
>> the allocation of interrupts and queues and have some sysfs knob to shut
>> down queues at runtime. If that shutdown results in releasing the queue
>> interrupt (via free_irq()) then the vector exhaustion problem goes away.
>>
>> Needs more thought and information (for network oblivious folks like
>> me).
>
> One piece of information that may be useful is that even tho the RX
> packets may be spread semi-randomly the user space can still control
> which queues are included in the mechanism. There is an indirection
> table in the HW which allows to weigh queues differently, or exclude
> selected queues from the spreading. Other mechanisms exist to filter
> flows onto specific queues.
>
> IOW just because a core has an queue/interrupt does not mean that
> interrupt will ever fire, provided its excluded from RSS.
>
> Another piece is that by default we suggest drivers allocate 8 RX
> queues, and online_cpus TX queues. The number of queues can be
> independently controlled via ethtool -L. Drivers which can't support
> separate queues will default to online_cpus queue pairs, and let
> ethtool -L only set the "combined" parameter.
>
I know the Intel drivers usually have defaulted to trying to maintain
queue pairs. I do not believe this is technically a HW restriction, but
it is heavily built into the way the drivers work today.
> There are drivers which always allocate online_cpus interrupts,
> and then some of them will go unused if #qs < #cpus.
>
>
Right.
> My unpopular opinion is that for networking devices all the heuristics
> we may come up with are going to be a dead end. We need an explicit API
> to allow users placing queues on cores, and use managed IRQs for data
> queues. (I'm assuming that managed IRQs will let us reliably map a MSI-X
> vector to a core :))
>
I don't think it is that unpopular... This is the direction I'd like to
see us go as well.
next prev parent reply other threads:[~2020-10-26 22:52 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-09-28 18:35 [PATCH v4 0/4] isolation: limit msix vectors to housekeeping CPUs Nitesh Narayan Lal
2020-09-28 18:35 ` [PATCH v4 1/4] sched/isolation: API to get number of " Nitesh Narayan Lal
2020-09-28 18:35 ` [PATCH v4 2/4] sched/isolation: Extend nohz_full to isolate managed IRQs Nitesh Narayan Lal
2020-10-23 13:25 ` Peter Zijlstra
2020-10-23 13:29 ` Frederic Weisbecker
2020-10-23 13:57 ` Nitesh Narayan Lal
2020-10-23 13:45 ` Nitesh Narayan Lal
2020-09-28 18:35 ` [PATCH v4 3/4] i40e: Limit msix vectors to housekeeping CPUs Nitesh Narayan Lal
2020-09-28 18:35 ` [PATCH v4 4/4] PCI: Limit pci_alloc_irq_vectors() " Nitesh Narayan Lal
2020-09-28 21:59 ` Bjorn Helgaas
2020-09-29 17:46 ` Christoph Hellwig
2020-10-16 12:20 ` Peter Zijlstra
2020-10-18 18:14 ` Nitesh Narayan Lal
2020-10-19 11:11 ` Peter Zijlstra
2020-10-19 14:00 ` Marcelo Tosatti
2020-10-19 14:25 ` Nitesh Narayan Lal
2020-10-20 7:30 ` Peter Zijlstra
2020-10-20 13:00 ` Nitesh Narayan Lal
2020-10-20 13:41 ` Peter Zijlstra
2020-10-20 14:39 ` Nitesh Narayan Lal
2020-10-22 17:47 ` Nitesh Narayan Lal
2020-10-23 8:58 ` Peter Zijlstra
2020-10-23 13:10 ` Nitesh Narayan Lal
2020-10-23 21:00 ` Thomas Gleixner
2020-10-26 13:35 ` Nitesh Narayan Lal
2020-10-26 13:57 ` Thomas Gleixner
2020-10-26 17:30 ` Marcelo Tosatti
2020-10-26 19:00 ` Thomas Gleixner
2020-10-26 19:11 ` Marcelo Tosatti
2020-10-26 19:21 ` Jacob Keller
2020-10-26 20:11 ` Thomas Gleixner
2020-10-26 21:11 ` Jacob Keller
2020-10-26 21:50 ` Thomas Gleixner
2020-10-26 22:13 ` Jakub Kicinski
2020-10-26 22:46 ` Thomas Gleixner
2020-10-26 22:52 ` Jacob Keller [this message]
2020-10-26 22:22 ` Nitesh Narayan Lal
2020-10-26 22:49 ` Thomas Gleixner
2020-10-26 23:08 ` Jacob Keller
2020-10-27 14:28 ` Thomas Gleixner
2020-10-27 11:47 ` Marcelo Tosatti
2020-10-27 14:43 ` Thomas Gleixner
2020-10-19 14:21 ` Frederic Weisbecker
2020-10-20 14:16 ` Thomas Gleixner
2020-10-20 16:18 ` Nitesh Narayan Lal
2020-10-20 18:07 ` Thomas Gleixner
2020-10-21 20:25 ` Thomas Gleixner
2020-10-21 21:04 ` Nitesh Narayan Lal
2020-10-22 0:02 ` Jakub Kicinski
2020-10-22 0:27 ` Jacob Keller
2020-10-22 8:28 ` Thomas Gleixner
2020-10-22 12:28 ` Marcelo Tosatti
2020-10-22 22:39 ` Thomas Gleixner
2020-10-01 15:49 ` [PATCH v4 0/4] isolation: limit msix vectors " Frederic Weisbecker
2020-10-08 21:40 ` Nitesh Narayan Lal
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:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=63c3484d-327e-5f37-7860-3af277c26711@intel.com \
--to=jacob.e.keller@intel.com \
--cc=bhelgaas@google.com \
--cc=dennis.dalessandro@intel.com \
--cc=frederic@kernel.org \
--cc=hch@infradead.org \
--cc=helgaas@kernel.org \
--cc=intel-wired-lan@lists.osuosl.org \
--cc=jeffrey.t.kirsher@intel.com \
--cc=jesse.brandeburg@intel.com \
--cc=jiri@nvidia.com \
--cc=jlelli@redhat.com \
--cc=juri.lelli@redhat.com \
--cc=kuba@kernel.org \
--cc=lgoncalv@redhat.com \
--cc=lihong.yang@intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pci@vger.kernel.org \
--cc=mike.marciniszyn@intel.com \
--cc=mingo@redhat.com \
--cc=mtosatti@redhat.com \
--cc=netdev@vger.kernel.org \
--cc=nitesh@redhat.com \
--cc=peterz@infradead.org \
--cc=sassmann@redhat.com \
--cc=tglx@linutronix.de \
--cc=thomas.lendacky@amd.com \
--cc=vincent.guittot@linaro.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* 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).