linux-kernel - Re: [PATCH net v3] net: ti: icssg-prueth: Fix 1 PPS sync

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: <7c3318f4-a2d4-4cbf-8a93-33c6a8afd6c4@ti.com>
Date: Mon, 4 Nov 2024 16:55:46 +0530
From: "Malladi, Meghana" <m-malladi@...com>
To: Jakub Kicinski <kuba@...nel.org>
CC: <vigneshr@...com>, <grygorii.strashko@...com>, <horms@...nel.org>,
        <jan.kiszka@...mens.com>, <diogo.ivo@...mens.com>, <pabeni@...hat.com>,
        <edumazet@...gle.com>, <davem@...emloft.net>, <andrew+netdev@...n.ch>,
        <linux-kernel@...r.kernel.org>, <netdev@...r.kernel.org>,
        <linux-arm-kernel@...ts.infradead.org>, <srk@...com>,
        Roger Quadros
	<rogerq@...nel.org>, <danishanwar@...com>,
        Vadim Fedorenko
	<vadim.fedorenko@...ux.dev>
Subject: Re: [PATCH net v3] net: ti: icssg-prueth: Fix 1 PPS sync



On 11/1/2024 7:29 AM, Jakub Kicinski wrote:
> On Mon, 28 Oct 2024 16:40:52 +0530 Meghana Malladi wrote:
>> The first PPS latch time needs to be calculated by the driver
>> (in rounded off seconds) and configured as the start time
>> offset for the cycle. After synchronizing two PTP clocks
>> running as master/slave, missing this would cause master
>> and slave to start immediately with some milliseconds
>> drift which causes the PPS signal to never synchronize with
>> the PTP master.
> 
> You're reading a 64b value in chunks, is it not possible that it'd wrap
> in between reads? This can be usually detected by reading high twice and
> making sure it didn't change.
> 
> Please fix or explain in the commit message why this is not a problem..
Yes I agree that there might be a wrap if the read isn't atomic. As 
suggested by Andrew I am currently not using custom read where I can 
implement the logic you suggested (reading high twice and making sure if 
didn't change). Can you share me some references where this logic is 
implemented in the kernel, so I can directly use that instead of writing 
custom functions.

Regards,
Meghana.