netdev - Re: [PATCH bpf-next v4 07/17] libbpf: Support drivers with non-combined channels

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: <20190614191236.2e498632@cakuba.netronome.com>
Date:   Fri, 14 Jun 2019 19:12:36 -0700
From:   Jakub Kicinski <jakub.kicinski@...ronome.com>
To:     Maxim Mikityanskiy <maximmi@...lanox.com>
Cc:     Alexei Starovoitov <ast@...nel.org>,
        Daniel Borkmann <daniel@...earbox.net>,
        Björn Töpel <bjorn.topel@...el.com>,
        Magnus Karlsson <magnus.karlsson@...el.com>,
        "bpf@...r.kernel.org" <bpf@...r.kernel.org>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        "David S. Miller" <davem@...emloft.net>,
        Saeed Mahameed <saeedm@...lanox.com>,
        Jonathan Lemon <bsd@...com>,
        Tariq Toukan <tariqt@...lanox.com>,
        Martin KaFai Lau <kafai@...com>,
        Song Liu <songliubraving@...com>, Yonghong Song <yhs@...com>,
        Maciej Fijalkowski <maciejromanfijalkowski@...il.com>
Subject: Re: [PATCH bpf-next v4 07/17] libbpf: Support drivers with
 non-combined channels

On Fri, 14 Jun 2019 13:25:05 +0000, Maxim Mikityanskiy wrote:
> Imagine you have configured the NIC to have the maximum supported amount 
> of channels. Then your formula in ethtool.c returns some value. Exactly 
> the same value should also be returned from libbpf's 
> xsk_get_max_queues(). It's achieved by applying your formula directly to 
> max.

I'm just trying to limit people inventing their own interpretations 
of this API.  Broadcom for instance does something dumb with current
counts I think they return curr.combined = curr.rx, even though there
is only curr.combined rings...

You will over allocate space for all NICs with return both combined and
non-combined counts.  But that's not a huge deal, not worth arguing about.
Moving on..