All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jesse Brandeburg <jesse.brandeburg@intel.com>
To: intel-wired-lan@osuosl.org
Subject: [Intel-wired-lan] [PATCH net-next v3 1/3] ice: update dim usage and moderation
Date: Mon, 20 Sep 2021 12:02:37 -0700	[thread overview]
Message-ID: <7ebc8fa6-79d0-13a3-e07a-247bfe8d8bcd@intel.com> (raw)
In-Reply-To: <d87087e8-51d7-4fd2-d594-a79b126fa14c@molgen.mpg.de>

On 9/19/2021 11:46 PM, Paul Menzel wrote:
> Dear Jesse,

Thanks for reviewing, I applied your text changes to v4.

> Am 24.08.21 um 03:12 schrieb Jesse Brandeburg:
>> Add a new helper for setting rate limit, which will be used more
>> in a followup patch.
> 
> Nice work. Could you please add, how to run the ping-pong test, so
> others can verify the behavior?

The test ran 32 rounds of netperf -t TCP_RR with a single thread, with
sizes of bytes transferred varying in powers of two from 1-64K.
each command was something like (for example for the 8K ping pong)
netperf -t TCP_RR -H host -l 30 -- -r 8K


>> @@ -1328,15 +1345,14 @@ static u32 ice_buildreg_itr(u16 itr_idx, u16 itr)
>> ? }
>> ? ? /**
>> - * ice_update_ena_itr - Update ITR moderation and re-enable MSI-X
>> interrupt
>> + * ice_enable_interrupt - re-enable MSI-X interrupt
>> ?? * @q_vector: the vector associated with the interrupt to enable
>> ?? *
>> - * Update the net_dim() algorithm and re-enable the interrupt
>> associated with
>> - * this vector.
>> - *
>> - * If the VSI is down, the interrupt will not be re-enabled.
>> + * If the VSI is down, the interrupt will not be re-enabled. Also,
>> + * when enabling the interrupt always reset the wb_on_itr to false
>> + * and trigger a software interrupt to clean out internal state.
> 
> Is the internal state cleaning required in the spec?

It is mentioned in the interrupt management section that triggering a
software interrupt will "flush" internal state. The spec just doesn't
mention when you should use that trick.


  reply	other threads:[~2021-09-20 19:02 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-08-24  1:12 [Intel-wired-lan] [PATCH net-next v3 0/3] ice: interrupt moderation updates Jesse Brandeburg
2021-08-24  1:12 ` [Intel-wired-lan] [PATCH net-next v3 1/3] ice: update dim usage and moderation Jesse Brandeburg
2021-09-20  5:09   ` G, GurucharanX
2021-09-20  6:46   ` Paul Menzel
2021-09-20 19:02     ` Jesse Brandeburg [this message]
2021-08-24  1:12 ` [Intel-wired-lan] [PATCH net-next v3 2/3] ice: fix rate limit update after coalesce change Jesse Brandeburg
2021-09-20  5:10   ` G, GurucharanX
2021-08-24  1:12 ` [Intel-wired-lan] [PATCH net-next v3 3/3] ice: fix software generating extra interrupts Jesse Brandeburg
2021-09-20  5:11   ` G, GurucharanX

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=7ebc8fa6-79d0-13a3-e07a-247bfe8d8bcd@intel.com \
    --to=jesse.brandeburg@intel.com \
    --cc=intel-wired-lan@osuosl.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 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.