netdev - Re: [PATCH v2 net-next] net/mlx5e: Report rx_discards_phy via rx_fifo

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: <20241123185024.21315d99@kernel.org>
Date: Sat, 23 Nov 2024 18:50:24 -0800
From: Jakub Kicinski <kuba@...nel.org>
To: Gal Pressman <gal@...dia.com>
Cc: Saeed Mahameed <saeed@...nel.org>, Saeed Mahameed <saeedm@...dia.com>,
 Yafang Shao <laoar.shao@...il.com>, ttoukan.linux@...il.com,
 tariqt@...dia.com, leon@...nel.org, netdev@...r.kernel.org,
 linux-rdma@...r.kernel.org
Subject: Re: [PATCH v2 net-next] net/mlx5e: Report rx_discards_phy via
 rx_fifo_errors

On Wed, 20 Nov 2024 08:04:35 +0200 Gal Pressman wrote:
> > The comment just says not to add what's already counted in missed,
> > because profcs adds the two and we'd end up double counting.  
> 
> So this is a procfs thing only?
> Does that mean that netlink's rx_dropped might be different than procfs'
> rx_dropped?

Yes, procfs and rtnl show "different" stats.
For more context on why I put the comment there -- some stats 
the drivers are supposed to fold (error stats from memory).

Legacy stats are tricky, it'd be a major review time investment 
to try to improve them..