All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jakub Kicinski <kuba@kernel.org>
To: "Nelson, Shannon" <shannon.nelson@amd.com>
Cc: netdev@vger.kernel.org, davem@davemloft.net, edumazet@google.com,
	pabeni@redhat.com, brett.creeley@amd.com, drivers@pensando.io
Subject: Re: [PATCH net-next 3/8] ionic: add private workqueue per-device
Date: Thu, 13 Jun 2024 15:19:28 -0700	[thread overview]
Message-ID: <20240613151928.6cc91d18@kernel.org> (raw)
In-Reply-To: <54c39843-b81b-4692-a22e-d2c51e617219@amd.com>

On Thu, 13 Jun 2024 13:38:18 -0700 Nelson, Shannon wrote:
> > little jobs little point of having your own wq, no?
> > At this point of reading the series its a bit unclear why
> > the wq separation is needed.  
> 
> Yes, when using only a single PF or two this doesn't look so bad to be 
> left on the system workqueue.  But we have a couple customers that want 
> to scale out to lots of VFs with multiple queues per VF, which 
> multiplies out 100's of queues getting workitems.  We thought that 
> instead of firebombing the system workqueue with a lot of little jobs, 
> we would give the scheduler a chance to work with our stuff separately, 
> and setting it up by device seemed like an easy enough way to partition 
> the work.  Other options might be one ionic wq used by all devices, or 
> maybe a wq per PF family?  Do you have a preference, or do you still 
> think that the system wq is enough?

No, no, code is fine. I was just complaining about the commit message :)
The math for how many work items you can see in reasonable scenarios
would be helpful to see when judging the need.

  reply	other threads:[~2024-06-13 22:19 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-06-10 23:06 [PATCH net-next 0/8] ionic: rework fix for doorbell miss Shannon Nelson
2024-06-10 23:06 ` [PATCH net-next 1/8] ionic: remove missed doorbell per-queue timer Shannon Nelson
2024-06-10 23:07 ` [PATCH net-next 2/8] ionic: Keep interrupt affinity up to date Shannon Nelson
2024-06-10 23:07 ` [PATCH net-next 3/8] ionic: add private workqueue per-device Shannon Nelson
2024-06-13  1:08   ` Jakub Kicinski
2024-06-13 20:38     ` Nelson, Shannon
2024-06-13 22:19       ` Jakub Kicinski [this message]
2024-06-13 22:40         ` Nelson, Shannon
2024-06-10 23:07 ` [PATCH net-next 4/8] ionic: add work item for missed-doorbell check Shannon Nelson
2024-06-13  1:19   ` Jakub Kicinski
2024-06-13 20:38     ` Nelson, Shannon
2024-06-14  1:26       ` Przemek Kitszel
2024-06-10 23:07 ` [PATCH net-next 5/8] ionic: add per-queue napi_schedule for doorbell check Shannon Nelson
2024-06-10 23:07 ` [PATCH net-next 6/8] ionic: check for queue deadline in doorbell_napi_work Shannon Nelson
2024-06-10 23:07 ` [PATCH net-next 7/8] ionic: Use an u16 for rx_copybreak Shannon Nelson
2024-06-15 21:20   ` David Laight
2024-06-16  1:28     ` Andrew Lunn
2024-06-16 10:27       ` David Laight
2024-06-17 16:24     ` Brett Creeley
2024-06-18  8:16       ` David Laight
2024-06-18 16:11         ` Brett Creeley
2024-06-10 23:07 ` [PATCH net-next 8/8] ionic: Only run the doorbell workaround for certain asic_type Shannon Nelson

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=20240613151928.6cc91d18@kernel.org \
    --to=kuba@kernel.org \
    --cc=brett.creeley@amd.com \
    --cc=davem@davemloft.net \
    --cc=drivers@pensando.io \
    --cc=edumazet@google.com \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=shannon.nelson@amd.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.