[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ZuR0ClWpxEk2I179@mini-arch>
Date: Fri, 13 Sep 2024 10:19:06 -0700
From: Stanislav Fomichev <stfomichev@...il.com>
To: Mina Almasry <almasrymina@...gle.com>
Cc: Stanislav Fomichev <sdf@...ichev.me>, netdev@...r.kernel.org,
davem@...emloft.net, edumazet@...gle.com, kuba@...nel.org,
pabeni@...hat.com
Subject: Re: [PATCH net-next 12/13] selftests: ncdevmem: Move ncdevmem under
drivers/net
On 09/13, Mina Almasry wrote:
> On Thu, Sep 12, 2024 at 10:13 AM Stanislav Fomichev <sdf@...ichev.me> wrote:
> >
> > This is where all the tests that depend on the HW functionality live in
>
> Is this true? My impression is that selftests/net verifies core
> functionality and drivers/net verifies more lower level driver
> (specific?) functionality. There are tests in selftests/net that
> depend on HW functionality, I think like bpf_offload.py.
>
> devmem tcp in my mind is primarily a core functionality and a lot of
> effort was put into making the driver bits of it as minimal as
> possible. Is there a need to move it or is this preference?
I'm wrong and it is actually drivers/net/hw, not drivers/net. But yes,
the driver/net/hw is the directory for the tests that depend on some
HW functionality (NIPA doesn't run the tests in this directory; but
the vendors eventually will). And until/if we get to loopback mode [1],
there is no way to run ncdevmem without special HW :-(
1: https://lore.kernel.org/netdev/20240913150913.1280238-5-sdf@fomichev.me/T/#m4fc52463857c19ad432db656c433cffdf74477a4
Powered by blists - more mailing lists