linux-kernel - Re: [PATCH net-next v2] ptp: add Alibaba CIPU PTP clock driver

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: <ea85f778-a2d4-439c-abbd-2a8ecea0e928@linux.alibaba.com>
Date: Mon, 30 Jun 2025 19:23:49 +0800
From: Wen Gu <guwen@...ux.alibaba.com>
To: Vadim Fedorenko <vadim.fedorenko@...ux.dev>, Andrew Lunn <andrew@...n.ch>
Cc: richardcochran@...il.com, andrew+netdev@...n.ch, davem@...emloft.net,
 edumazet@...gle.com, kuba@...nel.org, pabeni@...hat.com,
 xuanzhuo@...ux.alibaba.com, dust.li@...ux.alibaba.com,
 netdev@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH net-next v2] ptp: add Alibaba CIPU PTP clock driver



On 2025/6/27 18:59, Vadim Fedorenko wrote:
> On 27/06/2025 08:57, Andrew Lunn wrote:
> 
>>> +static int ptp_cipu_enable(struct ptp_clock_info *info,
>>> +               struct ptp_clock_request *request, int on)
>>> +{
>>> +    return -EOPNOTSUPP;
>>> +}
>>> +
>>> +static int ptp_cipu_settime(struct ptp_clock_info *p,
>>> +                const struct timespec64 *ts)
>>> +{
>>> +    return -EOPNOTSUPP;
>>> +}
>>> +
>>> +static int ptp_cipu_adjfine(struct ptp_clock_info *ptp, long scaled_ppm)
>>> +{
>>> +    return -EOPNOTSUPP;
>>> +}
>>> +
>>> +static int ptp_cipu_adjtime(struct ptp_clock_info *ptp, s64 delta)
>>> +{
>>> +    return -EOPNOTSUPP;
>>> +}
>>
>> I've not looked at the core. Are these actually required? Or if they
>> are missing, does the core default to -EOPNOTSUPP?
>>
> 
> I was going to say that these are not needed because posix clocks do
> check if callbacks are assigned and return -EOPNOTSUPP if they are not.
> That's why ptp_clock_* functions do call these callbacks without checks.

Hi Vadim, do you mean posix clock functions like this:

e.g. posix-clock.c:

static int pc_clock_settime(clockid_t id, const struct timespec64 *ts)
{
<...>
	if (cd.clk->ops.clock_settime)
		err = cd.clk->ops.clock_settime(cd.clk, ts);
	else
		err = -EOPNOTSUPP;
<...>
}

In ptp_clock.c, ops.clock_settime() is assigned to ptp_clock_settime(),
and it will call ptp->info->settime64() without checks. So I think these
'return -EOPNOTSUPP' functions are needed. Did I miss something?

Thanks!