[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20230119210842.5faf1e44@kernel.org>
Date: Thu, 19 Jan 2023 21:08:42 -0800
From: Jakub Kicinski <kuba@...nel.org>
To: Rahul Rameshbabu <rrameshbabu@...dia.com>
Cc: Jacob Keller <jacob.e.keller@...el.com>,
Saeed Mahameed <saeed@...nel.org>,
"David S. Miller" <davem@...emloft.net>,
Paolo Abeni <pabeni@...hat.com>,
Eric Dumazet <edumazet@...gle.com>,
Saeed Mahameed <saeedm@...dia.com>, <netdev@...r.kernel.org>,
Tariq Toukan <tariqt@...dia.com>,
Gal Pressman <gal@...dia.com>,
Richard Cochran <richardcochran@...il.com>,
Vincent Cheng <vincent.cheng.xh@...esas.com>
Subject: Re: [net-next 03/15] net/mlx5: Add adjphase function to support
hardware-only offset control
On Thu, 19 Jan 2023 20:26:04 -0800 Rahul Rameshbabu wrote:
> One of my concerns with doing this is breaking userspace expectations.
> In linuxptp, there is a configuration setting "write_phase_mode" and an
> expectation that when adjphase is called, there will not be a fallback
> to adjtime. This because adjphase is used in situations where small fine
> tuning is explicitly needed, so the errors would indicate a logical or
> situational error.
I don't mean fallback - just do what you do in mlx5 directly in
the core. The driver already does:
if delta < MAX
use precise method
else
use coarse method
> Quoting Vincent Cheng, the author of the adjphase functionality in the
> ptp core stack.
>
> -----BEGIN QUOTE-----
> adjtime modifies HW counter with a value to move the 1 PPS abruptly to new location.
> adjphase modifies the frequency to quickly nudge the 1 PPS to new location and also includes a HW filter to smooth out the adjustments and fine tune frequency.
>
> Continuous small offset adjustments using adjtime, likley see sudden shifts of the 1 PPS. The 1 PPS probably disappears and re-appears.
> Continuous small offset adjustments using adjphase, should see continuous 1 PPS.
>
> adjtime is good for large offset corrections
> adjphase is good for small offset corrections to allow HW filter to control the frequency instead of relying on SW filter.
Hm, so are you saying that:
adjtime(delta):
clock += delta
but:
adjfreq(delta):
on clock tick & while delta > 0:
clock += small_value
delta -= small_value
because from looking at mlx5 driver code its unclear whether the
implementation does a precise but one shot adjustment or gradual
adjustments.
Powered by blists - more mailing lists