[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAHk-=wgG=XFsHgMhgZpOM-M-PMa1cuz5=jExFv0KbajJ4JXN9w@mail.gmail.com>
Date: Sun, 21 Dec 2025 16:17:42 -0800
From: Linus Torvalds <torvalds@...ux-foundation.org>
To: Jakub Kicinski <kuba@...nel.org>, Tim Hostetler <thostet@...gle.com>,
Richard Cochran <richardcochran@...il.com>, Vadim Fedorenko <vadim.fedorenko@...ux.dev>,
Harshitha Ramamurthy <hramamurthy@...gle.com>, Kuniyuki Iwashima <kuniyu@...gle.com>,
Miri Korenblit <miriam.rachel.korenblit@...el.com>
Cc: davem@...emloft.net, netdev@...r.kernel.org, linux-kernel@...r.kernel.org,
pabeni@...hat.com
Subject: Re: [GIT PULL] Networking for Linux 6.19
On Tue, 2 Dec 2025 at 15:49, Jakub Kicinski <kuba@...nel.org> wrote:
>
> Tim Hostetler (5):
> ptp: Return -EINVAL on ptp_clock_register if required ops are NULL
I didn't notice until now, but this seems to result in a warning for
me on my old threadripper system:
info->n_alarm > PTP_MAX_ALARMS || (!info->gettimex64 &&
!info->gettime64) || !info->settime64
WARNING: drivers/ptp/ptp_clock.c:327 at
ptp_clock_register+0x33/0x6e0, CPU#0: NetworkManager/2370
Call Trace:
iwl_mvm_ptp_init+0xe6/0x160 [iwlmvm]
...
and the reason I didn't notice earlier is that it has no other ill
effects outside of the big ugly warning in the kernel messages.
Looking at the iwlwifi driver, it looks like the reason is that it
doesn't have a 'settime' operation, only an 'adjtime' one.
I can't tell whether it's the iwlwifi driver that should be fixed, or
whether that ptp warning condition is just bad. So I'm sending this to
the usual suspects, and hope somebody more competent can make that
judgement call.
Linus
Powered by blists - more mailing lists