public inbox for dev@dpdk.org
 help / color / mirror / Atom feed
From: scott.k.mitch1@gmail.com
To: dev@dpdk.org
Cc: stephen@networkplumber.org, Scott <scott.k.mitch1@gmail.com>
Subject: [PATCH v2 0/4] af_packet correctness, performance, cksum
Date: Wed, 28 Jan 2026 01:36:03 -0800	[thread overview]
Message-ID: <20260128093607.62908-1-scott.k.mitch1@gmail.com> (raw)
In-Reply-To: <20260127181355.98437-1-scott.k.mitch1@gmail.com>

From: Scott <scott.k.mitch1@gmail.com>

This series fixes critical thread safety bugs in the af_packet PMD
and adds performance optimizations.

Patch 1 fixes two major correctness issues:
- Thread safety: tp_status was accessed without memory barriers,
  violating the kernel's PACKET_MMAP protocol. On aarch64 and other
  weakly-ordered architectures, this causes packet corruption due to
  missing memory ordering. The fix matches the kernel's memory model:
  volatile unaligned reads/writes with explicit rte_smp_rmb/wmb
  barriers and __may_alias__ protection.

- Frame calculations: Fixed incorrect frame overhead and address
  calculations that caused memory corruption when frames don't evenly
  divide blocks.

Patches 2-4 add performance improvements:
- Patch 2: Bulk mbuf freeing, unlikely annotations, and prefetching
- Patch 3: TX poll control to reduce syscall overhead
- Patch 4: Software checksum offload support with shared rte_net
  utility

v2 changes:
- Patch 1: Rewrote to use volatile + barriers instead of C11 atomics
  to match kernel's memory model. Added dependency on patch-160274
  for __rte_may_alias attribute.
- Patch 4: Refactored to use shared rte_net_ip_udptcp_cksum_mbuf()
  utility function, eliminating code duplication with tap driver.

Scott Mitchell (4):
  net/af_packet: fix thread safety and frame calculations
  net/af_packet: RX/TX unlikely, bulk free, prefetch
  net/af_packet: tx poll control
  net/af_packet: software checksum

 doc/guides/nics/af_packet.rst             |   6 +-
 drivers/net/af_packet/rte_eth_af_packet.c | 257 ++++++++++++++++------
 drivers/net/tap/rte_eth_tap.c             |  61 +----
 lib/net/rte_net.h                         |  90 ++++++++
 4 files changed, 283 insertions(+), 131 deletions(-)

--
2.39.5 (Apple Git-154)


  parent reply	other threads:[~2026-01-28  9:36 UTC|newest]

Thread overview: 65+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-01-27 18:13 [PATCH v1 0/3] net/af_packet: correctness fixes and improvements scott.k.mitch1
2026-01-27 18:13 ` [PATCH v1 1/3] net/af_packet: fix thread safety and frame calculations scott.k.mitch1
2026-01-27 18:39   ` Stephen Hemminger
2026-01-28  1:35     ` Scott Mitchell
2026-01-27 18:13 ` [PATCH v1 2/3] net/af_packet: RX/TX rte_memcpy, bulk free, prefetch scott.k.mitch1
2026-01-27 18:54   ` Stephen Hemminger
2026-01-28  1:23     ` Scott Mitchell
2026-01-28  9:49       ` Morten Brørup
2026-01-28 15:37         ` Scott Mitchell
2026-01-28 16:57           ` Stephen Hemminger
2026-01-27 18:13 ` [PATCH v1 3/3] net/af_packet: software checksum and tx poll control scott.k.mitch1
2026-01-27 18:57   ` Stephen Hemminger
2026-01-28  7:05     ` Scott Mitchell
2026-01-28 17:36       ` Stephen Hemminger
2026-01-28 18:59         ` Scott Mitchell
2026-01-27 20:45   ` [REVIEW] " Stephen Hemminger
2026-01-28  9:36 ` scott.k.mitch1 [this message]
2026-01-28  9:36   ` [PATCH v2 1/4] net/af_packet: fix thread safety and frame calculations scott.k.mitch1
2026-01-28 16:59     ` Stephen Hemminger
2026-01-28 18:00       ` Scott Mitchell
2026-01-28 18:28         ` Stephen Hemminger
2026-01-28  9:36   ` [PATCH v2 2/4] net/af_packet: RX/TX unlikely, bulk free, prefetch scott.k.mitch1
2026-01-28  9:36   ` [PATCH v2 3/4] net/af_packet: tx poll control scott.k.mitch1
2026-01-28  9:36   ` [PATCH v2 4/4] net/af_packet: software checksum scott.k.mitch1
2026-01-28 18:27     ` Stephen Hemminger
2026-01-28 19:08       ` Scott Mitchell
2026-01-28 19:10   ` [PATCH v3 0/4] af_packet correctness, performance, cksum scott.k.mitch1
2026-01-28 19:10     ` [PATCH v3 1/4] net/af_packet: fix thread safety and frame calculations scott.k.mitch1
2026-01-28 19:10     ` [PATCH v3 2/4] net/af_packet: RX/TX unlikely, bulk free, prefetch scott.k.mitch1
2026-01-29  1:07       ` Stephen Hemminger
2026-02-02  5:29         ` Scott Mitchell
2026-01-28 19:10     ` [PATCH v3 3/4] net/af_packet: tx poll control scott.k.mitch1
2026-01-28 19:10     ` [PATCH v3 4/4] net/af_packet: software checksum scott.k.mitch1
2026-01-28 21:57       ` [REVIEW] " Stephen Hemminger
2026-02-02  7:55         ` Scott Mitchell
2026-02-02 16:58           ` Stephen Hemminger
2026-02-02  8:14     ` [PATCH v4 0/4] af_packet correctness, performance, cksum scott.k.mitch1
2026-02-02  8:14       ` [PATCH v4 1/4] net/af_packet: fix thread safety and frame calculations scott.k.mitch1
2026-02-02  8:14       ` [PATCH v4 2/4] net/af_packet: RX/TX bulk free, unlikely hint scott.k.mitch1
2026-02-02  8:14       ` [PATCH v4 3/4] net/af_packet: tx poll control scott.k.mitch1
2026-02-02  8:14       ` [PATCH v4 4/4] net/af_packet: add software checksum offload support scott.k.mitch1
2026-02-02 17:00         ` Stephen Hemminger
2026-02-02 18:47         ` Stephen Hemminger
2026-02-03  6:41           ` Scott Mitchell
2026-02-02 18:53       ` [PATCH v4 0/4] af_packet correctness, performance, cksum Stephen Hemminger
2026-02-03  7:07       ` [PATCH v5 " scott.k.mitch1
2026-02-03  7:07         ` [PATCH v5 1/4] net/af_packet: fix thread safety and frame calculations scott.k.mitch1
2026-02-03  7:07         ` [PATCH v5 2/4] net/af_packet: RX/TX bulk free, unlikely hint scott.k.mitch1
2026-02-03  7:07         ` [PATCH v5 3/4] net/af_packet: tx poll control scott.k.mitch1
2026-02-03  7:07         ` [PATCH v5 4/4] net/af_packet: add software checksum offload support scott.k.mitch1
2026-02-03  8:20           ` Scott Mitchell
2026-02-03 14:12             ` Stephen Hemminger
2026-02-04  2:59               ` Scott Mitchell
2026-02-03 14:13           ` Stephen Hemminger
2026-02-04  1:39             ` Scott Mitchell
2026-02-05 21:27               ` Stephen Hemminger
2026-02-06  1:11         ` [PATCH v6 0/4] af_packet correctness, performance, cksum scott.k.mitch1
2026-02-06  1:11           ` [PATCH v6 1/4] net/af_packet: fix thread safety and frame calculations scott.k.mitch1
2026-02-06  1:11           ` [PATCH v6 2/4] net/af_packet: RX/TX bulk free, unlikely hint scott.k.mitch1
2026-02-06  1:11           ` [PATCH v6 3/4] net/af_packet: tx poll control scott.k.mitch1
2026-02-06  1:11           ` [PATCH v6 4/4] net/af_packet: add software checksum offload support scott.k.mitch1
2026-02-06  1:49           ` [PATCH v6 0/4] af_packet correctness, performance, cksum Stephen Hemminger
2026-02-06  4:45             ` Scott Mitchell
2026-02-06 14:36             ` Morten Brørup
2026-02-06 16:11               ` Stephen Hemminger

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=20260128093607.62908-1-scott.k.mitch1@gmail.com \
    --to=scott.k.mitch1@gmail.com \
    --cc=dev@dpdk.org \
    --cc=stephen@networkplumber.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