[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <17c1c03a-1703-44e2-b586-01be0f86c0f7@redhat.com>
Date: Thu, 26 Sep 2024 12:14:44 +0200
From: Paolo Abeni <pabeni@...hat.com>
To: Sahand Evs <sahandevs@...il.com>, Eric Dumazet <edumazet@...gle.com>
Cc: Jordan Rife <jrife@...gle.com>, davem@...emloft.net, kuba@...nel.org,
netdev@...r.kernel.org
Subject: Re: [PATCH] net: expose __sock_sendmsg() symbol
On 9/23/24 21:36, Sahand Evs wrote:
> On Mon, Sep 23, 2024 at 10:30 PM Eric Dumazet <edumazet@...gle.com> wrote:
>> Convention on netdev mailing list is to not top post.
>>
>> Removing the static is not enough, a compiler and linker can
>> completely inline / delete this function.
>>
>> Anyway, I do not think sock_sendmsg() was part of any ABI.
[...]
>> Removing the static is not enough
> Should I also add a EXPORT_SYMBOL?
No, the relevant part is that the kernel don't guarantee
__sock_sendmsg()/sock_sendmsg() being constant (or present) across
different kernel versions (it's not part of any uAPI/ABI contract)
The user-space has to cope with that, trying to attach to different
function, according to what is available and visible. Possibly you don't
see __sock_sendmsg because you don't have the kernel debug infos
available. Try install them for the running kernel.
Cheers,
Paolo
Powered by blists - more mailing lists