All of lore.kernel.org
 help / color / mirror / Atom feed
From: Sean Young <sean@mess.org>
To: Viktor Prutyanov <viktor.prutyanov@phystech.edu>
Cc: mchehab@kernel.org, robh+dt@kernel.org, khilman@baylibre.com,
	narmstrong@baylibre.com, jbrunet@baylibre.com,
	martin.blumenstingl@googlemail.com, linux-media@vger.kernel.org,
	devicetree@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org,
	linux-amlogic@lists.infradead.org, rockosov@gmail.com
Subject: Re: [PATCH v5 2/2] media: rc: introduce Meson IR TX driver
Date: Fri, 16 Jul 2021 09:24:05 +0100	[thread overview]
Message-ID: <20210716082405.GA30719@gofer.mess.org> (raw)
In-Reply-To: <20210716013652.248bce6f@192.168.1.3>

Hi Viktor,

On Fri, Jul 16, 2021 at 01:36:52AM +0300, Viktor Prutyanov wrote:
> Hi Sean,
> 
> On Thu, 15 Jul 2021 22:40:01 +0100
> Sean Young <sean@mess.org> wrote:
> 
> > On Thu, Jul 15, 2021 at 12:27:06AM +0300, Viktor Prutyanov wrote:
 > > +	meson_irtx_fill_buf(ir, tx_buf, buf, len);
> > > +	dev_dbg(ir->dev, "TX buffer filled, length = %u\n", len);
> > > +
> > > +	spin_lock_irqsave(&ir->lock, flags);
> > > +	meson_irtx_update_buf(ir, tx_buf, len, 0);
> > > +	reinit_completion(&ir->completion);
> > > +	meson_irtx_send_buffer(ir);
> > > +	spin_unlock_irqrestore(&ir->lock, flags);
> > > +
> > > +	ret = wait_for_completion_interruptible(&ir->completion);
> > > +	dev_dbg(ir->dev, "TX %s\n", ret ? "interrupted" :
> > > "completed");  
> > 
> > Here two things can happen. One is, the process received a signal
> > (e.g. ^C). The other is that the hardware didn't issue any interrupts
> > due some problem somewhere. In the latter case, we only escape this
> > wait_for_completion_interruptable() when the user gets fed up and
> > presses ^C or something like that.
> > 
> > > +
> > > +	spin_lock_irqsave(&ir->lock, flags);
> > > +	kfree(ir->buf);
> > > +	meson_irtx_update_buf(ir, NULL, 0, 0);
> > > +	spin_unlock_irqrestore(&ir->lock, flags);  
> > 
> > Now it is possible that the buffer gets cleared before that IR was
> > sent, if the signal was received early enough. This means not all the
> > Tx was completed.
> > 
> > > +
> > > +	return len;  
> > 
> > Yet, we always return success.
> > 
> > In case no interrupts were generated we should return an error in a
> > timely manner, so the wait_for_completion() needs the timeout. You
> > can use the fact that the IR is never longer IR_MAX_DURATION (half a
> > second currently). Not sure what the returned error should be, maybe
> > -ETIMEDOUT?
> 
> As for me, ETIMEDOUT is OK.
> > 
> > The problem with the interruptable wait is that a process can receive
> > a signal at any time, and now when this happens your IR gets
> > truncated. I don't think this is what you want.
> 
> Should I replace wait_for_completion_interruptible by
> wait_for_completion_timeout in order to wait in uninterruptible way?

Yes, the process can receive a signal if the terminal is resized
(SIGWINCH) or if the process is backgrounded and then foregrounded with
^Z and fg (SIGCONT). If this happens during tx then the tx might be
incomplete.

Thanks,

Sean

WARNING: multiple messages have this Message-ID
From: Sean Young <sean@mess.org>
To: Viktor Prutyanov <viktor.prutyanov@phystech.edu>
Cc: mchehab@kernel.org, robh+dt@kernel.org, khilman@baylibre.com,
	narmstrong@baylibre.com, jbrunet@baylibre.com,
	martin.blumenstingl@googlemail.com, linux-media@vger.kernel.org,
	devicetree@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org,
	linux-amlogic@lists.infradead.org, rockosov@gmail.com
Subject: Re: [PATCH v5 2/2] media: rc: introduce Meson IR TX driver
Date: Fri, 16 Jul 2021 09:24:05 +0100	[thread overview]
Message-ID: <20210716082405.GA30719@gofer.mess.org> (raw)
In-Reply-To: <20210716013652.248bce6f@192.168.1.3>

Hi Viktor,

On Fri, Jul 16, 2021 at 01:36:52AM +0300, Viktor Prutyanov wrote:
> Hi Sean,
> 
> On Thu, 15 Jul 2021 22:40:01 +0100
> Sean Young <sean@mess.org> wrote:
> 
> > On Thu, Jul 15, 2021 at 12:27:06AM +0300, Viktor Prutyanov wrote:
 > > +	meson_irtx_fill_buf(ir, tx_buf, buf, len);
> > > +	dev_dbg(ir->dev, "TX buffer filled, length = %u\n", len);
> > > +
> > > +	spin_lock_irqsave(&ir->lock, flags);
> > > +	meson_irtx_update_buf(ir, tx_buf, len, 0);
> > > +	reinit_completion(&ir->completion);
> > > +	meson_irtx_send_buffer(ir);
> > > +	spin_unlock_irqrestore(&ir->lock, flags);
> > > +
> > > +	ret = wait_for_completion_interruptible(&ir->completion);
> > > +	dev_dbg(ir->dev, "TX %s\n", ret ? "interrupted" :
> > > "completed");  
> > 
> > Here two things can happen. One is, the process received a signal
> > (e.g. ^C). The other is that the hardware didn't issue any interrupts
> > due some problem somewhere. In the latter case, we only escape this
> > wait_for_completion_interruptable() when the user gets fed up and
> > presses ^C or something like that.
> > 
> > > +
> > > +	spin_lock_irqsave(&ir->lock, flags);
> > > +	kfree(ir->buf);
> > > +	meson_irtx_update_buf(ir, NULL, 0, 0);
> > > +	spin_unlock_irqrestore(&ir->lock, flags);  
> > 
> > Now it is possible that the buffer gets cleared before that IR was
> > sent, if the signal was received early enough. This means not all the
> > Tx was completed.
> > 
> > > +
> > > +	return len;  
> > 
> > Yet, we always return success.
> > 
> > In case no interrupts were generated we should return an error in a
> > timely manner, so the wait_for_completion() needs the timeout. You
> > can use the fact that the IR is never longer IR_MAX_DURATION (half a
> > second currently). Not sure what the returned error should be, maybe
> > -ETIMEDOUT?
> 
> As for me, ETIMEDOUT is OK.
> > 
> > The problem with the interruptable wait is that a process can receive
> > a signal at any time, and now when this happens your IR gets
> > truncated. I don't think this is what you want.
> 
> Should I replace wait_for_completion_interruptible by
> wait_for_completion_timeout in order to wait in uninterruptible way?

Yes, the process can receive a signal if the terminal is resized
(SIGWINCH) or if the process is backgrounded and then foregrounded with
^Z and fg (SIGCONT). If this happens during tx then the tx might be
incomplete.

Thanks,

Sean

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

WARNING: multiple messages have this Message-ID
From: Sean Young <sean@mess.org>
To: Viktor Prutyanov <viktor.prutyanov@phystech.edu>
Cc: mchehab@kernel.org, robh+dt@kernel.org, khilman@baylibre.com,
	narmstrong@baylibre.com, jbrunet@baylibre.com,
	martin.blumenstingl@googlemail.com, linux-media@vger.kernel.org,
	devicetree@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org,
	linux-amlogic@lists.infradead.org, rockosov@gmail.com
Subject: Re: [PATCH v5 2/2] media: rc: introduce Meson IR TX driver
Date: Fri, 16 Jul 2021 09:24:05 +0100	[thread overview]
Message-ID: <20210716082405.GA30719@gofer.mess.org> (raw)
In-Reply-To: <20210716013652.248bce6f@192.168.1.3>

Hi Viktor,

On Fri, Jul 16, 2021 at 01:36:52AM +0300, Viktor Prutyanov wrote:
> Hi Sean,
> 
> On Thu, 15 Jul 2021 22:40:01 +0100
> Sean Young <sean@mess.org> wrote:
> 
> > On Thu, Jul 15, 2021 at 12:27:06AM +0300, Viktor Prutyanov wrote:
 > > +	meson_irtx_fill_buf(ir, tx_buf, buf, len);
> > > +	dev_dbg(ir->dev, "TX buffer filled, length = %u\n", len);
> > > +
> > > +	spin_lock_irqsave(&ir->lock, flags);
> > > +	meson_irtx_update_buf(ir, tx_buf, len, 0);
> > > +	reinit_completion(&ir->completion);
> > > +	meson_irtx_send_buffer(ir);
> > > +	spin_unlock_irqrestore(&ir->lock, flags);
> > > +
> > > +	ret = wait_for_completion_interruptible(&ir->completion);
> > > +	dev_dbg(ir->dev, "TX %s\n", ret ? "interrupted" :
> > > "completed");  
> > 
> > Here two things can happen. One is, the process received a signal
> > (e.g. ^C). The other is that the hardware didn't issue any interrupts
> > due some problem somewhere. In the latter case, we only escape this
> > wait_for_completion_interruptable() when the user gets fed up and
> > presses ^C or something like that.
> > 
> > > +
> > > +	spin_lock_irqsave(&ir->lock, flags);
> > > +	kfree(ir->buf);
> > > +	meson_irtx_update_buf(ir, NULL, 0, 0);
> > > +	spin_unlock_irqrestore(&ir->lock, flags);  
> > 
> > Now it is possible that the buffer gets cleared before that IR was
> > sent, if the signal was received early enough. This means not all the
> > Tx was completed.
> > 
> > > +
> > > +	return len;  
> > 
> > Yet, we always return success.
> > 
> > In case no interrupts were generated we should return an error in a
> > timely manner, so the wait_for_completion() needs the timeout. You
> > can use the fact that the IR is never longer IR_MAX_DURATION (half a
> > second currently). Not sure what the returned error should be, maybe
> > -ETIMEDOUT?
> 
> As for me, ETIMEDOUT is OK.
> > 
> > The problem with the interruptable wait is that a process can receive
> > a signal at any time, and now when this happens your IR gets
> > truncated. I don't think this is what you want.
> 
> Should I replace wait_for_completion_interruptible by
> wait_for_completion_timeout in order to wait in uninterruptible way?

Yes, the process can receive a signal if the terminal is resized
(SIGWINCH) or if the process is backgrounded and then foregrounded with
^Z and fg (SIGCONT). If this happens during tx then the tx might be
incomplete.

Thanks,

Sean

_______________________________________________
linux-amlogic mailing list
linux-amlogic@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-amlogic

  reply	other threads:[~2021-07-16  8:24 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-07-14 21:27 [PATCH v5 0/2] media: rc: add support for Amlogic Meson IR blaster Viktor Prutyanov
2021-07-14 21:27 ` Viktor Prutyanov
2021-07-14 21:27 ` Viktor Prutyanov
2021-07-14 21:27 ` [PATCH v5 1/2] media: rc: meson-ir-tx: document device tree bindings Viktor Prutyanov
2021-07-14 21:27   ` Viktor Prutyanov
2021-07-14 21:27   ` Viktor Prutyanov
2021-07-14 21:27 ` [PATCH v5 2/2] media: rc: introduce Meson IR TX driver Viktor Prutyanov
2021-07-14 21:27   ` Viktor Prutyanov
2021-07-14 21:27   ` Viktor Prutyanov
2021-07-15 21:40   ` Sean Young
2021-07-15 21:40     ` Sean Young
2021-07-15 21:40     ` Sean Young
2021-07-15 22:36     ` Viktor Prutyanov
2021-07-15 22:36       ` Viktor Prutyanov
2021-07-15 22:36       ` Viktor Prutyanov
2021-07-16  8:24       ` Sean Young [this message]
2021-07-16  8:24         ` Sean Young
2021-07-16  8:24         ` Sean Young

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=20210716082405.GA30719@gofer.mess.org \
    --to=sean@mess.org \
    --cc=devicetree@vger.kernel.org \
    --cc=jbrunet@baylibre.com \
    --cc=khilman@baylibre.com \
    --cc=linux-amlogic@lists.infradead.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-media@vger.kernel.org \
    --cc=martin.blumenstingl@googlemail.com \
    --cc=mchehab@kernel.org \
    --cc=narmstrong@baylibre.com \
    --cc=robh+dt@kernel.org \
    --cc=rockosov@gmail.com \
    --cc=viktor.prutyanov@phystech.edu \
    --subject='Re: [PATCH v5 2/2] media: rc: introduce Meson IR TX driver' \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.