netdev - Re: [PATCH net-next 3/3] eth: bnxt: handle invalid Tx completions more gracefully

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

Message-ID: <20230711212421.02a36ab3@kernel.org>
Date: Tue, 11 Jul 2023 21:24:21 -0700
From: Jakub Kicinski <kuba@...nel.org>
To: Michael Chan <michael.chan@...adcom.com>
Cc: davem@...emloft.net, netdev@...r.kernel.org, edumazet@...gle.com,
 pabeni@...hat.com
Subject: Re: [PATCH net-next 3/3] eth: bnxt: handle invalid Tx completions
 more gracefully

On Tue, 11 Jul 2023 21:11:02 -0700 Michael Chan wrote:
> These are per ring counters.  During complete reset, we free the ring
> and allocate a new ring.  So re-applying these counters to the new
> ring doesn't make sense because the new ring is not necessarily the
> same as the old one.  So I think we'll need to have a total for each
> of these and we'll save the snapshot of the total before reset and
> restore the snapshot after reset.  Does that make sense?

Not entirely sure what you mean by total. The counters are reported 
in ethool -S per ring and (aggregated) in ip -s -s.

Are you saying that in ethtool -S in addition to per-ring counts
we'd report a "total" which is sum(current per ring) + saved?
If so - that makes sense, yup.

ip -s -s sort of follows so no need discussing.

You mention reset but the errors counters should survive close() /
open() cycles as well.