From: Aleksandr Loktionov <aleksandr.loktionov@intel.com>
To: intel-wired-lan@lists.osuosl.org, anthony.l.nguyen@intel.com,
aleksandr.loktionov@intel.com
Cc: netdev@vger.kernel.org
Subject: [PATCH iwl-next 0/4] ixgbe: adaptive ITR tuning to prevent RX starvation and ACK overdrive
Date: Tue, 15 Sep 2026 14:55:34 +0200 [thread overview]
Message-ID: <20260915125538.3975870-1-aleksandr.loktionov@intel.com> (raw)
This series was split out of a larger internal "ixgbe: nits and
improvements" cleanup batch. It tunes the adaptive-ITR algorithm and is
kept as four small, independently reviewable/revertible patches rather
than one monolithic rewrite:
1. Lower IXGBE_ITR_ADAPTIVE_MAX_USECS from 126 to 84 so the minimum
bulk-mode interrupt rate cannot drop low enough to starve the Rx
descriptor ring under sustained full-line-rate traffic.
2. Add an ixgbe_container_is_rx() helper and give Rx containers their
own latency thresholds, replacing the shared Tx/Rx heuristic that
the lowered maximum would otherwise interact with poorly.
3. Limit how far a single ITR update can decrease while in latency
mode, so a low-rate ACK-only stream cannot overdrive the interrupt
rate in one step.
4. Name the usecs mask used by ixgbe_set_itr() instead of relying on
an open-coded complement of the mode-flag bit.
A fifth patch from the same original batch, removing a redundant
ixgbe_ping_all_vfs() call from the link watchdog handlers, addresses an
unrelated VF mailbox race rather than ITR/RX-starvation behavior and is
sent separately so this series stays focused on one topic.
Alexander Duyck (4):
ixgbe: lower IXGBE_ITR_ADAPTIVE_MAX_USECS to prevent RX starvation
ixgbe: add ixgbe_container_is_rx() helper and refine RX adaptive ITR
ixgbe: limit ITR decrease in latency mode to prevent ACK overdrive
ixgbe: add IXGBE_ITR_ADAPTIVE_MASK_USECS constant
drivers/net/ethernet/intel/ixgbe/ixgbe.h | 3 +-
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 80 +++++++++++--------
2 files changed, 49 insertions(+), 34 deletions(-)
--
2.52.0
next reply other threads:[~2026-09-15 12:55 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-09-15 12:55 Aleksandr Loktionov [this message]
2026-09-15 12:55 ` [PATCH iwl-next 1/4] ixgbe: lower IXGBE_ITR_ADAPTIVE_MAX_USECS to prevent RX starvation Aleksandr Loktionov
2026-09-16 14:41 ` Simon Horman
2026-09-15 12:55 ` [PATCH iwl-next 2/4] ixgbe: add ixgbe_container_is_rx() helper and refine RX adaptive ITR Aleksandr Loktionov
2026-09-16 14:41 ` Simon Horman
2026-09-15 12:55 ` [PATCH iwl-next 3/4] ixgbe: limit ITR decrease in latency mode to prevent ACK overdrive Aleksandr Loktionov
2026-09-15 12:55 ` [PATCH iwl-next 4/4] ixgbe: add IXGBE_ITR_ADAPTIVE_MASK_USECS constant Aleksandr Loktionov
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=20260915125538.3975870-1-aleksandr.loktionov@intel.com \
--to=aleksandr.loktionov@intel.com \
--cc=anthony.l.nguyen@intel.com \
--cc=intel-wired-lan@lists.osuosl.org \
--cc=netdev@vger.kernel.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