Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Subject: Re: [PATCH v5 3/5] counter: Add character device interface
To:     William Breathitt Gray <vilhelm.gray@gmail.com>, jic23@kernel.org
Cc:     kamel.bouhara@bootlin.com, gwendal@chromium.org,
        alexandre.belloni@bootlin.com, linux-iio@vger.kernel.org,
        linux-kernel@vger.kernel.org,
        linux-stm32@st-md-mailman.stormreply.com,
        linux-arm-kernel@lists.infradead.org, syednwaris@gmail.com,
        patrick.havelange@essensium.com, fabrice.gasnier@st.com,
        mcoquelin.stm32@gmail.com, alexandre.torgue@st.com
References: <cover.1601170670.git.vilhelm.gray@gmail.com>
 <00be1fccc672c5207f3b04fe4cc09c29e22641f4.1601170670.git.vilhelm.gray@gmail.com>
From:   David Lechner <david@lechnology.com>
Message-ID: <67a0290e-731b-822a-5113-30b56bde6c88@lechnology.com>
Date:   Wed, 14 Oct 2020 12:43:08 -0500
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101
 Thunderbird/68.10.0
MIME-Version: 1.0
In-Reply-To: <00be1fccc672c5207f3b04fe4cc09c29e22641f4.1601170670.git.vilhelm.gray@gmail.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 9/26/20 9:18 PM, William Breathitt Gray wrote:
> diff --git a/drivers/counter/counter-chrdev.c b/drivers/counter/counter-chrdev.c
> new file mode 100644
> index 000000000000..2be3846e4105
> --- /dev/null
> +++ b/drivers/counter/counter-chrdev.c


> +/**
> + * counter_push_event - queue event for userspace reading
> + * @counter:	pointer to Counter structure
> + * @event:	triggered event
> + * @channel:	event channel
> + *
> + * Note: If no one is watching for the respective event, it is silently
> + * discarded.
> + *
> + * RETURNS:
> + * 0 on success, negative error number on failure.
> + */
> +int counter_push_event(struct counter_device *const counter, const u8 event,
> +		       const u8 channel)
> +{
> +	struct counter_event ev = {0};
> +	unsigned int copied = 0;
> +	unsigned long flags;
> +	struct counter_event_node *event_node;
> +	struct counter_comp_node *comp_node;
> +	int err;
> +
> +	ev.timestamp = ktime_get_ns();
> +	ev.watch.event = event;
> +	ev.watch.channel = channel;
> +
> +	raw_spin_lock_irqsave(&counter->events_lock, flags);
> +
> +	/* Search for event in the list */
> +	list_for_each_entry(event_node, &counter->events_list, l)
> +		if (event_node->event == event &&
> +		    event_node->channel == channel)
> +			break;
> +
> +	/* If event is not in the list */
> +	if (&event_node->l == &counter->events_list)
> +		goto exit_early;
> +
> +	/* Read and queue relevant comp for userspace */
> +	list_for_each_entry(comp_node, &event_node->comp_list, l) {
> +		err = counter_get_data(counter, comp_node, &ev.value_u8);

Currently all counter devices are memory mapped devices so calling
counter_get_data() here with interrupts disabled is probably OK, but
if any counter drivers are added that use I2C/SPI/etc. that will take
a long time to read, it would cause problems leaving interrupts
disabled here.

Brainstorming: Would it make sense to separate the event from the
component value being read? As I mentioned in one of my previous
reviews, I think there are some cases where we would just want to
know when an event happened and not read any additional data anyway.
In the case of a slow communication bus, this would also let us
queue the event in the kfifo and notify poll right away and then
defer the reads in a workqueue for later.


> +		if (err)
> +			goto err_counter_get_data;
> +
> +		ev.watch.component = comp_node->component;
> +
> +		copied += kfifo_put(&counter->events, ev);
> +	}
> +
> +	if (copied)
> +		wake_up_poll(&counter->events_wait, EPOLLIN);
> +
> +exit_early:
> +	raw_spin_unlock_irqrestore(&counter->events_lock, flags);
> +
> +	return 0;
> +
> +err_counter_get_data:
> +	raw_spin_unlock_irqrestore(&counter->events_lock, flags);
> +	return err;
> +}