[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <87a5m03ne0.ffs@tglx>
Date: Wed, 10 Apr 2024 23:42:47 +0200
From: Thomas Gleixner <tglx@...utronix.de>
To: lakshmi.sowjanya.d@...el.com, jstultz@...gle.com, giometti@...eenne.com,
corbet@....net, linux-kernel@...r.kernel.org
Cc: x86@...nel.org, netdev@...r.kernel.org, linux-doc@...r.kernel.org,
intel-wired-lan@...ts.osuosl.org, andriy.shevchenko@...ux.intel.com,
eddie.dong@...el.com, christopher.s.hall@...el.com,
jesse.brandeburg@...el.com, davem@...emloft.net,
alexandre.torgue@...s.st.com, joabreu@...opsys.com,
mcoquelin.stm32@...il.com, perex@...ex.cz, linux-sound@...r.kernel.org,
anthony.l.nguyen@...el.com, peter.hilber@...nsynergy.com,
pandith.n@...el.com, subramanian.mohan@...el.com,
thejesh.reddy.t.r@...el.com, lakshmi.sowjanya.d@...el.com
Subject: Re: [PATCH v6 02/11] e1000e: remove convert_art_to_tsc()
On Wed, Apr 10 2024 at 17:18, lakshmi.sowjanya.d@...el.com wrote:
> From: Thomas Gleixner <tglx@...utronix.de>
>
> Remove convert_art_to_tsc() function call, Pass system clock cycles and
> clocksource ID as input to get_device_system_crosststamp().
This is wrong as this does not pass anything as input. The function is
called from get_device_system_crosststamp() and the data is passed back
via the system_counterval pointer. It also lacks context.
Something like this:
"The core code provides a mechanism to convert the ART base clock to
the corresponding TSC value without requiring an architecture
specific function.
All what is required is to store the ART clocksoure ID and the
cycles value in the provided system_counterval structure.
Replace the direct conversion via convert_art_to_tsc() by filling in
the required data.
No functional change intended."
Hmm?
Thanks,
tglx
Powered by blists - more mailing lists