Netdev List
 help / color / mirror / Atom feed
From: Satish Kharat <satishkh@cisco.com>
To: netdev-bot+sashiko@kernel.org
Cc: netdev@vger.kernel.org, sebaddel@cisco.com, aradhana@cisco.com,
	rfaucett@cisco.com, andrew+netdev@lunn.ch, davem@davemloft.net,
	edumazet@kernel.org, kuba@kernel.org, pabeni@redhat.com
Subject: Re: [PATCH net v7 1/2] enic: retain admin-channel DMA until queues stop
Date: Fri, 13 Nov 2026 07:17:22 -0800	[thread overview]
Message-ID: <20261113151722.3116868-1-satishkh@cisco.com> (raw)
In-Reply-To: <179450328635.2199510.2101319582570477640@kernel.org>

Thanks for the review.

> enic_admin_channel_close() can return an error after releasing the graph.

Agreed. The QP_DISABLE result and the direct queue-stop checks describe
different outcomes. If both complete disable helpers succeed, the resources
are released and the state becomes DOWN, even when firmware rejected
QP_DISABLE. The partial-open caller can therefore print a misleading
close-failure message. The queue-stop proof and resource handling remain
correct.

> The quarantine check can close a PF and prevent its reset.

The ownership check deliberately remains outside enic_has_admin_chan().
Disabling SR-IOV clears ENIC_SRIOV_ENABLED but can leave retained admin
resources with unresolved DMA ownership. Clearing that flag must not
bypass the reset guard.

I agree that PF reset and reopen policy need to be consistent. PF
admin-channel activation is dormant in this series because
.sriov_configure is not registered.

> A failed internal reopen can leave the interface marked up and allow
> repeated NAPI teardown.

Agreed. This is the pre-existing datapath lifecycle issue acknowledged
in my earlier replies. Adding dev_close() directly after failed reopen
would invoke another stop and can encounter the same problem.

I will handle the pre-existing datapath lifecycle issue in a separate
fix. The PF quarantine and recovery policy will be addressed before PF
SR-IOV activation, and I will follow up separately on the close-status
diagnostic.

If there are no further concerns with the two fixes in V7, would you
please consider applying the series to net?

Regards,
Satish

  reply	other threads:[~2026-11-13 15:18 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-09 20:39 [PATCH net v7 0/2] enic: retain admin DMA resources and identify V2 VFs Satish Kharat
2026-11-09 20:39 ` [PATCH net v7 1/2] enic: retain admin-channel DMA until queues stop Satish Kharat
2026-11-12 17:08   ` netdev-bot+sashiko
2026-11-13 15:17     ` Satish Kharat [this message]
2026-11-09 20:39 ` [PATCH net v7 2/2] enic: require DEVCMD2 and set the V2 VF mailbox source Satish Kharat
2026-11-16 23:30 ` [PATCH net v7 0/2] enic: retain admin DMA resources and identify V2 VFs patchwork-bot+netdevbpf

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=20261113151722.3116868-1-satishkh@cisco.com \
    --to=satishkh@cisco.com \
    --cc=andrew+netdev@lunn.ch \
    --cc=aradhana@cisco.com \
    --cc=davem@davemloft.net \
    --cc=edumazet@kernel.org \
    --cc=kuba@kernel.org \
    --cc=netdev-bot+sashiko@kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=rfaucett@cisco.com \
    --cc=sebaddel@cisco.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox