public inbox for netdev@vger.kernel.org
 help / color / mirror / Atom feed
From: Potin Lai <potin.lai.pt@gmail.com>
To: Samuel Mendoza-Jonas <sam@mendozajonas.com>,
	 "David S. Miller" <davem@davemloft.net>,
	Eric Dumazet <edumazet@google.com>,
	 Jakub Kicinski <kuba@kernel.org>,
	Paolo Abeni <pabeni@redhat.com>,  Simon Horman <horms@kernel.org>,
	Ivan Mikhaylov <fr0st61te@gmail.com>,
	 Paul Fertser <fercerpav@gmail.com>,
	Patrick Williams <patrick@stwcx.xyz>
Cc: netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
	 Cosmo Chou <cosmo.chou@quantatw.com>,
	Potin Lai <potin.lai@quantatw.com>,
	 Potin Lai <potin.lai.pt@gmail.com>,
	Cosmo Chou <chou.cosmo@gmail.com>
Subject: [PATCH v3 2/2] net/ncsi: fix state race during channel probe completion
Date: Mon, 13 Jan 2025 10:34:48 +0800	[thread overview]
Message-ID: <20250113-fix-ncsi-mac-v3-2-564c8277eb1d@gmail.com> (raw)
In-Reply-To: <20250113-fix-ncsi-mac-v3-0-564c8277eb1d@gmail.com>

From: Cosmo Chou <chou.cosmo@gmail.com>

During channel probing, the last NCSI_PKT_CMD_DP command can trigger
an unnecessary schedule_work() via ncsi_free_request(). We observed
that subsequent config states were triggered before the scheduled
work completed, causing potential state handling issues.

Fix this by clearing req_flags when processing the last package.

Fixes: 8e13f70be05e ("net/ncsi: Probe single packages to avoid conflict")
Signed-off-by: Cosmo Chou <chou.cosmo@gmail.com>
---
 net/ncsi/ncsi-manage.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/net/ncsi/ncsi-manage.c b/net/ncsi/ncsi-manage.c
index bf276eaf9330..99e3a5883617 100644
--- a/net/ncsi/ncsi-manage.c
+++ b/net/ncsi/ncsi-manage.c
@@ -1491,7 +1491,10 @@ static void ncsi_probe_channel(struct ncsi_dev_priv *ndp)
 		}
 		break;
 	case ncsi_dev_state_probe_dp:
-		ndp->pending_req_num = 1;
+		if (ndp->package_probe_id + 1 < 8)
+			ndp->pending_req_num = 1;
+		else
+			nca.req_flags = 0;
 
 		/* Deselect the current package */
 		nca.type = NCSI_PKT_CMD_DP;

-- 
2.31.1


  parent reply	other threads:[~2025-01-13  2:37 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-01-13  2:34 [PATCH v3 0/2] net/ncsi: fix oem gma command handling and state race issue Potin Lai
2025-01-13  2:34 ` [PATCH v3 1/2] net/ncsi: fix locking in Get MAC Address handling Potin Lai
2025-01-13  2:34 ` Potin Lai [this message]
2025-01-14 12:55   ` [PATCH v3 2/2] net/ncsi: fix state race during channel probe completion Paul Fertser
2025-01-16  0:38     ` Potin Lai
2025-01-16 15:29       ` [PATCH] net/ncsi: wait for the last response to Deselect Package before configuring channel Paul Fertser
2025-01-23  4:10         ` 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=20250113-fix-ncsi-mac-v3-2-564c8277eb1d@gmail.com \
    --to=potin.lai.pt@gmail.com \
    --cc=chou.cosmo@gmail.com \
    --cc=cosmo.chou@quantatw.com \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=fercerpav@gmail.com \
    --cc=fr0st61te@gmail.com \
    --cc=horms@kernel.org \
    --cc=kuba@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=patrick@stwcx.xyz \
    --cc=potin.lai@quantatw.com \
    --cc=sam@mendozajonas.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