[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <3c21a8909a745d5ae1bfa466aaac5c40752cdfef.camel@oss.nxp.com>
Date: Mon, 03 Jan 2022 14:28:43 +0200
From: "Radu Nicolae Pirea (NXP OSS)" <radu-nicolae.pirea@....nxp.com>
To: Richard Cochran <richardcochran@...il.com>
Cc: netdev@...r.kernel.org, linux-kernel@...r.kernel.org,
christian.herber@....com, andrew@...n.ch, hkallweit1@...il.com,
linux@...linux.org.uk, davem@...emloft.net, kuba@...nel.org
Subject: Re: [PATCH v2 2/2] phy: nxp-c45-tja11xx: read the tx timestamp
without lock
On Thu, 2021-12-23 at 12:24 -0800, Richard Cochran wrote:
> On Wed, Dec 22, 2021 at 11:34:53PM +0200, Radu Pirea (NXP OSS) wrote:
> > Reading the tx timestamps can be done in parallel with adjusting
> > the LTC
> > value.
> >
> > Calls to nxp_c45_get_hwtxts() are always serialised. If the phy
> > interrupt is enabled, .do_aux_work() will not call
> > nxp_c45_get_hwtxts.
>
> Reviewing the code, I see what you mean. However, the serialization
> is completely non-obvious, and future changes to the driver could
> easily spoil the implicit serialization. Given that the mutex is not
> highly contended, I suggest dropping this patch in order to
> future-proof the driver.
>
I agree. This patch can be dropped.
Radu P.
> Thanks,
> Richard
Powered by blists - more mailing lists