netdev - Re: [PATCH v2] ice: fix race condition in TX timestamp ring cleanup

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-Id: <20260206045426.3547398-1-kmta1236@gmail.com>
Date: Fri,  6 Feb 2026 13:54:26 +0900
From: Keita Morisaki <kmta1236@...il.com>
To: maciej.fijalkowski@...el.com
Cc: aleksandr.loktionov@...el.com,
	alice.michael@...el.com,
	andrew+netdev@...n.ch,
	davem@...emloft.net,
	edumazet@...gle.com,
	intel-wired-lan@...ts.osuosl.org,
	kmta1236@...il.com,
	kuba@...nel.org,
	netdev@...r.kernel.org,
	pabeni@...hat.com,
	paul.greenwalt@...el.com,
	przemyslaw.kitszel@...el.com,
	tony.nguyen@...el.com
Subject: Re: [PATCH v2] ice: fix race condition in TX timestamp ring cleanup

> am I reading this right or you only modified flags to be bitmap on Tx side
> and Rx is left as-is?
>
> Would be nice to reflect this on Rx side as well, but I know it is out of
> the scope for this patch.
>
> Would you like to do a follow-up patch for that which would be sent to
> -next tree?

Yes, I intentionally left RX side as is because it seems less obviously that
rx flags are touched concurrently than tx ones.

But sure, it's a good idea to make the handling more defensive on the rx side as well.
I think it can be a follow-up patch as TX has more obvious race condition.

> Also your patch should contain the tree you are dedicating your work:
> [PATCH net v2] ice: fix race condition in TX timestamp ring cleanup

Didn't know that! Thank you for the comment. I will follow this in future
(I'm also happy to resend the V2 patch with the updated subject if you think it's).

> As for rest of the stuff it makes sense to me, however best if Paul could
> give his ack here.