Received-SPF: pass (google.com: domain of linux-kernel+bounces-104451-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Message-ID: <63d6a4b0cb2155d5c194b3b65fa60c985a6338ca.camel@gmail.com>
Subject: Re: [PATCH v2 4/6] iio: pressure: Simplify and make more clear
 temperature readings
From: Angel Iglesias <ang.iglesiasg@gmail.com>
To: Vasileios Amoiridis <vassilisamir@gmail.com>
Cc: Jonathan Cameron <jic23@kernel.org>, lars@metafoo.de, 
 andriy.shevchenko@linux.intel.com, mazziesaccount@gmail.com,
 ak@it-klinger.de,  petre.rodan@subdimension.ro, linus.walleij@linaro.org,
 phil@raspberrypi.com,  579lpy@gmail.com, linux-iio@vger.kernel.org,
 linux-kernel@vger.kernel.org
Date: Fri, 15 Mar 2024 14:28:30 +0100
In-Reply-To: <ZfQPVlqv7A3zxExl@vamoirid-EDL-PC>
References: <20240313174007.1934983-1-vassilisamir@gmail.com>
	 <20240313174007.1934983-5-vassilisamir@gmail.com>
	 <20240314150959.585367b5@jic23-huawei>
	 <20240314201718.GD1964894@vamoiridPC>
	 <46389801aeb20f18affed86d979aff7a62cf36d5.camel@gmail.com>
	 <ZfQPVlqv7A3zxExl@vamoirid-EDL-PC>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
User-Agent: Evolution 3.50.4 (by Flathub.org) 
Precedence: bulk
MIME-Version: 1.0

On Fri, 2024-03-15 at 10:05 +0100, Vasileios Amoiridis wrote:
> On Fri, Mar 15, 2024 at 12:22:50AM +0100, Angel Iglesias wrote:
> > On Thu, 2024-03-14 at 21:17 +0100, Vasileios Amoiridis wrote:
> > > On Thu, Mar 14, 2024 at 03:09:59PM +0000, Jonathan Cameron wrote:
> > > > On Wed, 13 Mar 2024 18:40:05 +0100
> > > > Vasileios Amoiridis <vassilisamir@gmail.com> wrote:
> > > >=20
> > > > > The read_press/read_humid functions need the updated t_fine value
> > > > > in order to calculate the current pressure/humidity. Temperature
> > > > > reads should be removed from the read_press/read_humid functions
> > > > > and should be placed in the oneshot captures before the pressure
> > > > > and humidity reads. This makes the code more intuitive.
> > > > >=20
> > > > > Signed-off-by: Vasileios Amoiridis <vassilisamir@gmail.com>
> > > >=20
> > > > To me this makes the use of these calls less obvious than they were
> > > > previously.=C2=A0 The calls are made close to where t_fine is used =
and
> > > > don't have to go via the indirection of chip_info.
> > > >=20
> > > > So I disagree. I think this change makes the code a lot less
> > > > clear.
> > > >=20
> > >=20
> > > This was mainly driven by the fact that I wanted to avoid reading
> > > the temperature 3 times in case temp, press and humid are enabled
> > > and there are consecutive buffer readings. But thank you for the
> > > proposal I really appreciate it!
> > >=20
> >=20
> > Hi, just a side note reflecting on this. Depending on your sampling
> > frequency
> > and registers data shadowing, to avoid compensating with different samp=
les
> > between readings, we should be doing burst readings to get a bundle of =
the
> > temperature+pressure and/or humidity.
> > On the bmp/bme280 and bmp380 this can be done as registers are contiguo=
us on
> > the
> > memory. On the bmp580 this is not a problem as the values are already
> > compensated, you`ll get always the latest reading.
> >=20
> > Kind regard,
> > Angel
>=20
> Hi Angel,
>=20
> Thank you for pointing this out! Indeed that's true but I noticed that th=
is is
> not
> the case for the BMP{085/180} devices. I just feel that some changes migh=
t
> make
> data acquisition/processing faster for a device (like the one you propose=
d)
> but
> it might make the code much more unreadable and unmaintanable. I will try=
 and
> see if something could be done in that sense but I feel that keeping it s=
imple
> will
> be good for everyone!
>=20
> Cheers,
> Vasilis

Yeah, data adquisition on bmp085/180 is already different as they don't sup=
port
continuous mode as the newer models and you have to warm-up the sensor and =
do
one-shot readings. There's already a different code path in place for that
models. I guess that is the price to pay to support that wide range of
sensors...
Anyway, this patches are already big and you're doing quite a lot of heavy-
lifting right now, so don't pay much attention to my ramblings! Regardless,
happy to help with tasks polishing and updating this driver :)

Kind regards,
Angel