LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: "Sa, Nuno" <Nuno.Sa@analog.com>
To: Jonathan Cameron <jic23@kernel.org>,
	Miquel Raynal <miquel.raynal@bootlin.com>
Cc: Lars-Peter Clausen <lars@metafoo.de>,
	Thomas Petazzoni <thomas.petazzoni@bootlin.com>,
	"linux-iio@vger.kernel.org" <linux-iio@vger.kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: RE: [PATCH 14/16] iio: adc: max1027: Consolidate the end of conversion helper
Date: Mon, 6 Sep 2021 09:12:23 +0000	[thread overview]
Message-ID: <SA1PR03MB6355961592622ED9DF24064299D29@SA1PR03MB6355.namprd03.prod.outlook.com> (raw)
In-Reply-To: <20210905104126.156bcb59@jic23-huawei>



> -----Original Message-----
> From: Jonathan Cameron <jic23@kernel.org>
> Sent: Sunday, September 5, 2021 11:41 AM
> To: Miquel Raynal <miquel.raynal@bootlin.com>
> Cc: Sa, Nuno <Nuno.Sa@analog.com>; Lars-Peter Clausen
> <lars@metafoo.de>; Thomas Petazzoni
> <thomas.petazzoni@bootlin.com>; linux-iio@vger.kernel.org; linux-
> kernel@vger.kernel.org
> Subject: Re: [PATCH 14/16] iio: adc: max1027: Consolidate the end of
> conversion helper
> 
> [External]
> 
> On Fri, 3 Sep 2021 16:46:50 +0200
> Miquel Raynal <miquel.raynal@bootlin.com> wrote:
> 
> > Hi Nuno,
> >
> > Nuno.Sa@analog.com wrote on Fri, 3 Sep 2021 14:28:52 +0000:
> >
> > > Hi Miquel,
> > >
> > > > From: Miquel Raynal <miquel.raynal@bootlin.com>
> > > > Sent: Thursday, September 2, 2021 5:13 PM
> > > > To: Sa, Nuno <Nuno.Sa@analog.com>
> > > > Cc: Jonathan Cameron <jic23@kernel.org>; Lars-Peter Clausen
> > > > <lars@metafoo.de>; Thomas Petazzoni
> > > > <thomas.petazzoni@bootlin.com>; linux-iio@vger.kernel.org;
> linux-
> > > > kernel@vger.kernel.org
> > > > Subject: Re: [PATCH 14/16] iio: adc: max1027: Consolidate the
> end of
> > > > conversion helper
> > > >
> > > > Hi Nuno,
> > > >
> > > > "Sa, Nuno" <Nuno.Sa@analog.com> wrote on Mon, 30 Aug 2021
> > > > 12:44:48
> > > > +0000:
> > > >
> > > > > > -----Original Message-----
> > > > > > From: Jonathan Cameron <jic23@kernel.org>
> > > > > > Sent: Monday, August 30, 2021 12:37 PM
> > > > > > To: Miquel Raynal <miquel.raynal@bootlin.com>
> > > > > > Cc: Lars-Peter Clausen <lars@metafoo.de>; Thomas
> Petazzoni
> > > > > > <thomas.petazzoni@bootlin.com>; linux-iio@vger.kernel.org;
> > > > linux-
> > > > > > kernel@vger.kernel.org
> > > > > > Subject: Re: [PATCH 14/16] iio: adc: max1027: Consolidate the
> end
> > > > of
> > > > > > conversion helper
> > > > > >
> > > > > > On Wed, 18 Aug 2021 13:11:37 +0200
> > > > > > Miquel Raynal <miquel.raynal@bootlin.com> wrote:
> > > > > >
> > > > > > > Now that we have a dedicated handler for End Of
> Conversion
> > > > > > interrupts,
> > > > > > > let's create a second path:
> > > > > > > - Situation 1: we are using the external hardware trigger, a
> > > > > > conversion
> > > > > > >   has been triggered and the ADC pushed the data to its
> FIFO, we
> > > > > > need to
> > > > > > >   retrieve the data and push it to the IIO buffers.
> > > > > > > - Situation 2: we are not using the external hardware
> trigger,
> > > > hence
> > > > > > we
> > > > > > >   are likely waiting in a blocked thread waiting for this
> interrupt to
> > > > > > >   happen: in this case we just wake up the waiting thread.
> > > > > > >
> > > > > > > Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
> > > > > > > ---
> > > > > > >  drivers/iio/adc/max1027.c | 20 +++++++++++++++++---
> > > > > > >  1 file changed, 17 insertions(+), 3 deletions(-)
> > > > > > >
> > > > > > > diff --git a/drivers/iio/adc/max1027.c
> > > > b/drivers/iio/adc/max1027.c
> > > > > > > index 8d86e77fb5db..8c5995ae59f2 100644
> > > > > > > --- a/drivers/iio/adc/max1027.c
> > > > > > > +++ b/drivers/iio/adc/max1027.c
> > > > > > > @@ -235,6 +235,7 @@ struct max1027_state {
> > > > > > >  	struct iio_trigger		*trig;
> > > > > > >  	__be16				*buffer;
> > > > > > >  	struct mutex			lock;
> > > > > > > +	bool				data_rdy;
> > > > > > >  	bool				cnvst_trigger;
> > > > > > >  	u8				reg
> ____cacheline_aligned;
> > > > > > >  };
> > > > > > > @@ -243,12 +244,22 @@ static
> > > > > > DECLARE_WAIT_QUEUE_HEAD(max1027_queue);
> > > > > > >
> > > > > > >  static int max1027_wait_eoc(struct iio_dev *indio_dev)
> > > > > > >  {
> > > > > > > +	struct max1027_state *st = iio_priv(indio_dev);
> > > > > > >  	unsigned int conversion_time =
> > > > > > MAX1027_CONVERSION_UDELAY;
> > > > > > > +	int ret;
> > > > > > >
> > > > > > > -	if (indio_dev->active_scan_mask)
> > > > > > > -		conversion_time *= hweight32(*indio_dev-
> > > > > > >active_scan_mask);
> > > > > > > +	if (st->spi->irq) {
> > > > > > > +		ret =
> > > > > > wait_event_interruptible_timeout(max1027_queue,
> > > > > > > +						       st-
> >data_rdy, HZ /
> > > > > > 1000);
> > > > > > > +		st->data_rdy = false;
> > > > > > > +		if (ret == -ERESTARTSYS)
> > > > > > > +			return ret;
> > > > > > > +	} else {
> > > > > > > +		if (indio_dev->active_scan_mask)
> > > > > > > +			conversion_time *=
> hweight32(*indio_dev-
> > > > > > >active_scan_mask);
> > > > > > >
> > > > > > > -	usleep_range(conversion_time, conversion_time * 2);
> > > > > > > +		usleep_range(conversion_time,
> conversion_time * 2);
> > > > > > > +	}
> > > > > > >
> > > > > > >  	return 0;
> > > > > > >  }
> > > > > > > @@ -481,6 +492,9 @@ static irqreturn_t
> > > > > > max1027_eoc_irq_handler(int irq, void *private)
> > > > > > >  	if (st->cnvst_trigger) {
> > > > > > >  		ret = max1027_read_scan(indio_dev);
> > > > > > >  		iio_trigger_notify_done(indio_dev->trig);
> > > > > > > +	} else {
> > > > > > > +		st->data_rdy = true;
> > > > > > > +		wake_up(&max1027_queue);
> > > > > >
> > > > > > I can't see why a queue is appropriate for this.  Use a
> completion
> > > > and
> > > > > > have
> > > > > > one per instance of the device.  No need for the flag etc in
> that
> > > > case as
> > > > > > complete() means we have had an interrupt.
> > > > > >
> > > > >
> > > > > In the case that 'st-> cnvst_trigger' is not set but the spi IRQ
> > > > > is present, we will wait until we get 'wake_up()' called from
> here. I
> > > > wonder if
> > > > > that is a good idea as the device own trigger is not being used.
> FWIW,
> > > > I think this
> > > > > sync logic is a bit confusing... I would still use the normal trigger
> > > > infrastructure
> > > > > ('iio_trigger_generic_data_rdy_poll()') and use the
> 'cnvst_trigger'
> > > > flag in the
> > > > > trigger handler to manually start conversions + wait till eoc. But I
> > > > might be missing
> > > > > something though.
> > > >
> > > > I implemented it your way, but I think I found a situation that was
> not
> > > > fully handled (the 3rd), which makes the handler very
> complicated
> > > > as we need to handle all the following cases:
> > > > 1/ no trigger, irq enabled -> single read EOC interrupt
> > > > 2/ external trigger, no irq -> handle the whole conversion process
> > > > 3/ external trigger, irq enabled -> handle the whole conversion
> process
> > > >    but also have a dedicated condition to handle the EOC interrupt
> > > >    properly (fortunately this is a threaded handler that can be
> > > >    preempted): we need to wait from the handler itself that the
> > > >    handler gets called again: the first time it is executed as
> > > >    "pollfunc", the second time as "EOC interrupt". In the second
> > > >    instance, call complete() in order to deliver the first running
> > > >    instance of the handler and continue until the reading part.
> > > > 4/ cnvst trigger, irq enabled -> only reads the data.
> > > > 5/ cnvst trigger, irq disabled -> not possible.
> > > >
> > > > I added a lot of comments to make it clearer.
> > > >
> > > > > Regarding this handler, I just realized that this is the hard IRQ
> handler
> > > > which
> > > > > might end up calling 'max1027_read_scan()' which in turn calls
> > > > 'spi_read()'. Am I
> > > > > missing something here?
> > > >
> > > > I renamed it to make it clear, but this is already a threaded
> handler.
> > > >
> > >
> > > Hmm, I think I get what you're trying to do.... FWIW, I think you're
> just going
> > > into a lot of trouble here for scenario 3 (I assume external trigger is
> something
> > > else other the device own one). IMO, I would just assume that if
> we are using
> > > an external trigger we have to wait (sleep) for the end of
> conversion (i.e, I would
> > > not care about the IRQ in this case). It would make things much
> more simpler and
> > > I guess it should be expected that if some user is deliberately not
> using the
> > > device own trigger, will have to wait more for scans.
> >
> > I thought about that, but this means playing with enabling/disabling
> > IRQs, which in the end I fear could be almost as verbose :/
> >
> > Can you look at the new implementation that I proposed and give
> me your
> > feedback on it?
> > [PATCH v2 15/16] iio: adc: max1027: Add support for external triggers
> >
> > > I cannot also see a reason why someone would want to use some
> > > external trigger if the device one is available... Does it really make
> sense?
> >
> > I would say that "genericity" is a good enough reason for that, but in
> > practice I agree with you.
> 
> Synchronising capture across a number of devices is usual reason for
> this.  It's not perfect, but particularly if you have a device
> with a complex trigger that is not occurring on at an even 'tick' of
> time, then in those cases there isn't really any other way of doing

I see. That is indeed a valid use case for this...

> it.  However that irq is unrelated to this driver.  This driver
> sees a trigger which it then identifies as not being it's own and
> from that fires the conversion and waits for a completion.
> The EOC interrupt handler knows we are in this mode and calls
> complete() rather than iio_poll_trigger()
> 

Yes, I do understand the logic. I'm just not 100% convinced that
the added complexity for the syncing brings much added value. It
evens looks like that the IRQ is not being used for normal raw reads?
So, I would [probably] just act as no IRQ is present for external triggers
(I do not think it's not that bad to enable/disable the IRQ on the trigger
state cb). Alternatively, starting to us the IRQ for raw reads would
make a stronger argument to enforce this logic as it would be more
consistent...

But I guess this also boils down to personal preference so as long as you
are fine with this logic :) ...

- Nuno Sá

  reply	other threads:[~2021-09-06  9:12 UTC|newest]

Thread overview: 71+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-08-18 11:11 [PATCH 00/16] Bring software triggers support to MAX1027-like ADCs Miquel Raynal
2021-08-18 11:11 ` [PATCH 01/16] iio: adc: max1027: Fix wrong shift with 12-bit devices Miquel Raynal
2021-08-20  7:02   ` Sa, Nuno
2021-08-30  9:57     ` Jonathan Cameron
2021-08-18 11:11 ` [PATCH 02/16] iio: adc: max1027: Fix the number of max1X31 channels Miquel Raynal
2021-08-20  7:03   ` Sa, Nuno
2021-08-30 10:00     ` Jonathan Cameron
2021-08-18 11:11 ` [PATCH 03/16] iio: adc: max1027: Push only the requested samples Miquel Raynal
2021-08-20  7:10   ` Sa, Nuno
2021-08-30 10:07     ` Jonathan Cameron
2021-08-30 10:49       ` Sa, Nuno
2021-08-30 14:29         ` Jonathan Cameron
2021-08-30 15:02           ` Sa, Nuno
2021-09-01  8:12             ` Miquel Raynal
2021-09-04 14:06               ` Jonathan Cameron
2021-09-06  8:59               ` Sa, Nuno
2021-09-06 16:56                 ` Jonathan Cameron
2021-09-06 17:34                   ` Miquel Raynal
2021-08-30 10:06   ` Jonathan Cameron
2021-08-18 11:11 ` [PATCH 04/16] iio: adc: max1027: Lower conversion time Miquel Raynal
2021-08-20  7:12   ` Sa, Nuno
2021-08-30 10:10   ` Jonathan Cameron
2021-08-18 11:11 ` [PATCH 05/16] iio: adc: max1027: Drop extra warning message Miquel Raynal
2021-08-20  7:12   ` Sa, Nuno
2021-08-30 10:12   ` Jonathan Cameron
2021-08-18 11:11 ` [PATCH 06/16] iio: adc: max1027: Rename a helper Miquel Raynal
2021-08-20  7:13   ` Sa, Nuno
2021-08-18 11:11 ` [PATCH 07/16] iio: adc: max1027: Create a helper to configure the trigger Miquel Raynal
2021-08-20  7:16   ` Sa, Nuno
2021-08-30 10:16   ` Jonathan Cameron
2021-08-18 11:11 ` [PATCH 08/16] iio: adc: max1027: Explain better how the trigger state gets changed Miquel Raynal
2021-08-20  7:17   ` Sa, Nuno
2021-08-18 11:11 ` [PATCH 09/16] iio: adc: max1027: Create a helper to configure the channels to scan Miquel Raynal
2021-08-20  7:18   ` Sa, Nuno
2021-08-18 11:11 ` [PATCH 10/16] iio: adc: max1027: Prevent single channel accesses during buffer reads Miquel Raynal
2021-08-20  7:20   ` Sa, Nuno
2021-08-20  7:30     ` Sa, Nuno
2021-08-30 10:20       ` Jonathan Cameron
2021-09-02  8:56         ` Miquel Raynal
2021-08-18 11:11 ` [PATCH 11/16] iio: adc: max1027: Separate the IRQ handler from the read logic Miquel Raynal
2021-08-20  7:23   ` Sa, Nuno
2021-09-02  8:55     ` Miquel Raynal
2021-09-04 14:08       ` Jonathan Cameron
2021-08-18 11:11 ` [PATCH 12/16] iio: adc: max1027: Introduce an end of conversion helper Miquel Raynal
2021-08-20  7:28   ` Sa, Nuno
2021-09-02  9:26     ` Miquel Raynal
2021-08-30 10:34   ` Jonathan Cameron
2021-08-18 11:11 ` [PATCH 13/16] iio: adc: max1027: Prepare re-using the EOC interrupt Miquel Raynal
2021-08-20  7:31   ` Sa, Nuno
2021-08-30 10:30   ` Jonathan Cameron
2021-08-30 10:47   ` Jonathan Cameron
2021-08-18 11:11 ` [PATCH 14/16] iio: adc: max1027: Consolidate the end of conversion helper Miquel Raynal
2021-08-20  7:45   ` Sa, Nuno
2021-08-30 10:37   ` Jonathan Cameron
2021-08-30 12:44     ` Sa, Nuno
2021-08-30 14:32       ` Jonathan Cameron
2021-09-02 15:12       ` Miquel Raynal
2021-09-03 14:28         ` Sa, Nuno
2021-09-03 14:46           ` Miquel Raynal
2021-09-05  9:41             ` Jonathan Cameron
2021-09-06  9:12               ` Sa, Nuno [this message]
2021-09-06  9:30                 ` Sa, Nuno
2021-08-18 11:11 ` [PATCH 15/16] iio: adc: max1027: Support software triggers Miquel Raynal
2021-08-20  7:58   ` Sa, Nuno
2021-09-02 12:25     ` Miquel Raynal
2021-09-03 14:20       ` Sa, Nuno
2021-08-30 10:50   ` Jonathan Cameron
2021-09-02 15:21     ` Miquel Raynal
2021-09-05  9:43       ` Jonathan Cameron
2021-08-18 11:11 ` [PATCH 16/16] iio: adc: max1027: Enable software triggers to be used without IRQ Miquel Raynal
2021-08-30 10:54   ` Jonathan Cameron

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=SA1PR03MB6355961592622ED9DF24064299D29@SA1PR03MB6355.namprd03.prod.outlook.com \
    --to=nuno.sa@analog.com \
    --cc=jic23@kernel.org \
    --cc=lars@metafoo.de \
    --cc=linux-iio@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=miquel.raynal@bootlin.com \
    --cc=thomas.petazzoni@bootlin.com \
    --subject='RE: [PATCH 14/16] iio: adc: max1027: Consolidate the end of conversion helper' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).