From: Peter Hurley <peter-WaGBZJeGNqdsbIuE7sb01tBPR1lH4CV8@public.gmane.org>
To: Frans Klaver
<frans.klaver-MHHw4NDrbWwAvxtiuMwx3w@public.gmane.org>,
Tony Lindgren <tony-4v6yS6AI5VpBDgjK7y7TUQ@public.gmane.org>
Cc: Felipe Balbi <balbi-l0cyMroinI0@public.gmane.org>,
Rob Herring <robh+dt-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>,
Pawel Moll <pawel.moll-5wv7dgnIgG8@public.gmane.org>,
Mark Rutland <mark.rutland-5wv7dgnIgG8@public.gmane.org>,
Ian Campbell
<ijc+devicetree-KcIKpvwj1kUDXYZnReoRVg@public.gmane.org>,
Kumar Gala <galak-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org>,
Greg Kroah-Hartman
<gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org>,
Alexey Pelykh
<alexey.pelykh-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>,
Jiri Slaby <jslaby-AlSwsSmVLrQ@public.gmane.org>,
linux-serial-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-omap-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: Re: [PATCH 3/4] tty: omap-serial: use threaded interrupt handler
Date: Mon, 15 Sep 2014 13:31:56 -0400 [thread overview]
Message-ID: <5417228C.5070503@hurleysoftware.com> (raw)
In-Reply-To: <54170823.7040801-WaGBZJeGNqdsbIuE7sb01tBPR1lH4CV8@public.gmane.org>
On 09/15/2014 11:39 AM, Peter Hurley wrote:
> On 09/15/2014 10:00 AM, Frans Klaver wrote:
>> At 3.6Mbaud, with slightly over 2Mbit/s data coming in, we see 1600 uart
>> rx buffer overflows within 30 seconds. Threading the interrupt handling reduces
>> this to about 170 overflows in 10 minutes.
>
> Why is the threadirqs kernel boot option not sufficient?
> Or conversely, shouldn't this be selectable?
Also, do you see the same performance differential when you implement this
in the 8250 driver (that is, on top of Sebastian's omap->8250 conversion)?
> Regards,
> Peter Hurley
>
> PS - To overflow the 64 byte RX FIFO at those data rates means interrupt
> latency in excess of 250us?
>
>> In practice this therefore reduces the need for hardware flow control,
>> meaning the sending side doesn't have to buffer as much either.
>>
>> Signed-off-by: Frans Klaver <frans.klaver-MHHw4NDrbWwAvxtiuMwx3w@public.gmane.org>
>> ---
>> drivers/tty/serial/omap-serial.c | 30 +++++++++++++++++++++++-------
>> 1 file changed, 23 insertions(+), 7 deletions(-)
>>
>> diff --git a/drivers/tty/serial/omap-serial.c b/drivers/tty/serial/omap-serial.c
>> index 7d3f557..398139a 100644
>> --- a/drivers/tty/serial/omap-serial.c
>> +++ b/drivers/tty/serial/omap-serial.c
>> @@ -575,6 +575,20 @@ static void serial_omap_rdi(struct uart_omap_port *up, unsigned int lsr)
>> }
>>
>> /**
>> + * serial_omap_fast_irq() - schedule interrupt handling
>> + */
>> +static irqreturn_t serial_omap_fast_irq(int irq, void *dev_id)
>> +{
>> + struct uart_omap_port *up = dev_id;
>> + unsigned int iir = serial_in(up, UART_IIR);
>> +
>> + if (iir & UART_IIR_NO_INT)
>> + return IRQ_NONE;
>> +
>> + return IRQ_WAKE_THREAD;
>> +}
>> +
>> +/**
>> * serial_omap_irq() - This handles the interrupt from one port
>> * @irq: uart port irq number
>> * @dev_id: uart port info
>> @@ -584,7 +598,6 @@ static irqreturn_t serial_omap_irq(int irq, void *dev_id)
>> struct uart_omap_port *up = dev_id;
>> unsigned int iir, lsr;
>> unsigned int type;
>> - irqreturn_t ret = IRQ_NONE;
>> int max_count = 256;
>>
>> spin_lock(&up->port.lock);
>> @@ -595,7 +608,6 @@ static irqreturn_t serial_omap_irq(int irq, void *dev_id)
>> if (iir & UART_IIR_NO_INT)
>> break;
>>
>> - ret = IRQ_HANDLED;
>> lsr = serial_in(up, UART_LSR);
>>
>> /* extract IRQ type from IIR register */
>> @@ -634,7 +646,7 @@ static irqreturn_t serial_omap_irq(int irq, void *dev_id)
>> pm_runtime_put_autosuspend(up->dev);
>> up->port_activity = jiffies;
>>
>> - return ret;
>> + return IRQ_HANDLED;
>> }
>>
>> static unsigned int serial_omap_tx_empty(struct uart_port *port)
>> @@ -731,15 +743,19 @@ static int serial_omap_startup(struct uart_port *port)
>> /*
>> * Allocate the IRQ
>> */
>> - retval = request_irq(up->port.irq, serial_omap_irq, up->port.irqflags,
>> - up->name, up);
>> + retval = request_threaded_irq(up->port.irq, serial_omap_fast_irq,
>> + serial_omap_irq,
>> + IRQF_ONESHOT | up->port.irqflags,
>> + up->name, up);
>> if (retval)
>> return retval;
>>
>> /* Optional wake-up IRQ */
>> if (up->wakeirq) {
>> - retval = request_irq(up->wakeirq, serial_omap_irq,
>> - up->port.irqflags, up->name, up);
>> + retval = request_threaded_irq(up->wakeirq, serial_omap_fast_irq,
>> + serial_omap_irq,
>> + IRQF_ONESHOT | up->port.irqflags,
>> + up->name, up);
>> if (retval) {
>> free_irq(up->port.irq, up);
>> return retval;
>>
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
WARNING: multiple messages have this Message-ID (diff)
From: peter@hurleysoftware.com (Peter Hurley)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 3/4] tty: omap-serial: use threaded interrupt handler
Date: Mon, 15 Sep 2014 13:31:56 -0400 [thread overview]
Message-ID: <5417228C.5070503@hurleysoftware.com> (raw)
In-Reply-To: <54170823.7040801@hurleysoftware.com>
On 09/15/2014 11:39 AM, Peter Hurley wrote:
> On 09/15/2014 10:00 AM, Frans Klaver wrote:
>> At 3.6Mbaud, with slightly over 2Mbit/s data coming in, we see 1600 uart
>> rx buffer overflows within 30 seconds. Threading the interrupt handling reduces
>> this to about 170 overflows in 10 minutes.
>
> Why is the threadirqs kernel boot option not sufficient?
> Or conversely, shouldn't this be selectable?
Also, do you see the same performance differential when you implement this
in the 8250 driver (that is, on top of Sebastian's omap->8250 conversion)?
> Regards,
> Peter Hurley
>
> PS - To overflow the 64 byte RX FIFO at those data rates means interrupt
> latency in excess of 250us?
>
>> In practice this therefore reduces the need for hardware flow control,
>> meaning the sending side doesn't have to buffer as much either.
>>
>> Signed-off-by: Frans Klaver <frans.klaver@xsens.com>
>> ---
>> drivers/tty/serial/omap-serial.c | 30 +++++++++++++++++++++++-------
>> 1 file changed, 23 insertions(+), 7 deletions(-)
>>
>> diff --git a/drivers/tty/serial/omap-serial.c b/drivers/tty/serial/omap-serial.c
>> index 7d3f557..398139a 100644
>> --- a/drivers/tty/serial/omap-serial.c
>> +++ b/drivers/tty/serial/omap-serial.c
>> @@ -575,6 +575,20 @@ static void serial_omap_rdi(struct uart_omap_port *up, unsigned int lsr)
>> }
>>
>> /**
>> + * serial_omap_fast_irq() - schedule interrupt handling
>> + */
>> +static irqreturn_t serial_omap_fast_irq(int irq, void *dev_id)
>> +{
>> + struct uart_omap_port *up = dev_id;
>> + unsigned int iir = serial_in(up, UART_IIR);
>> +
>> + if (iir & UART_IIR_NO_INT)
>> + return IRQ_NONE;
>> +
>> + return IRQ_WAKE_THREAD;
>> +}
>> +
>> +/**
>> * serial_omap_irq() - This handles the interrupt from one port
>> * @irq: uart port irq number
>> * @dev_id: uart port info
>> @@ -584,7 +598,6 @@ static irqreturn_t serial_omap_irq(int irq, void *dev_id)
>> struct uart_omap_port *up = dev_id;
>> unsigned int iir, lsr;
>> unsigned int type;
>> - irqreturn_t ret = IRQ_NONE;
>> int max_count = 256;
>>
>> spin_lock(&up->port.lock);
>> @@ -595,7 +608,6 @@ static irqreturn_t serial_omap_irq(int irq, void *dev_id)
>> if (iir & UART_IIR_NO_INT)
>> break;
>>
>> - ret = IRQ_HANDLED;
>> lsr = serial_in(up, UART_LSR);
>>
>> /* extract IRQ type from IIR register */
>> @@ -634,7 +646,7 @@ static irqreturn_t serial_omap_irq(int irq, void *dev_id)
>> pm_runtime_put_autosuspend(up->dev);
>> up->port_activity = jiffies;
>>
>> - return ret;
>> + return IRQ_HANDLED;
>> }
>>
>> static unsigned int serial_omap_tx_empty(struct uart_port *port)
>> @@ -731,15 +743,19 @@ static int serial_omap_startup(struct uart_port *port)
>> /*
>> * Allocate the IRQ
>> */
>> - retval = request_irq(up->port.irq, serial_omap_irq, up->port.irqflags,
>> - up->name, up);
>> + retval = request_threaded_irq(up->port.irq, serial_omap_fast_irq,
>> + serial_omap_irq,
>> + IRQF_ONESHOT | up->port.irqflags,
>> + up->name, up);
>> if (retval)
>> return retval;
>>
>> /* Optional wake-up IRQ */
>> if (up->wakeirq) {
>> - retval = request_irq(up->wakeirq, serial_omap_irq,
>> - up->port.irqflags, up->name, up);
>> + retval = request_threaded_irq(up->wakeirq, serial_omap_fast_irq,
>> + serial_omap_irq,
>> + IRQF_ONESHOT | up->port.irqflags,
>> + up->name, up);
>> if (retval) {
>> free_irq(up->port.irq, up);
>> return retval;
>>
WARNING: multiple messages have this Message-ID (diff)
From: Peter Hurley <peter@hurleysoftware.com>
To: Frans Klaver <frans.klaver@xsens.com>, Tony Lindgren <tony@atomide.com>
Cc: Felipe Balbi <balbi@ti.com>, Rob Herring <robh+dt@kernel.org>,
Pawel Moll <pawel.moll@arm.com>,
Mark Rutland <mark.rutland@arm.com>,
Ian Campbell <ijc+devicetree@hellion.org.uk>,
Kumar Gala <galak@codeaurora.org>,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
Alexey Pelykh <alexey.pelykh@gmail.com>,
Jiri Slaby <jslaby@suse.cz>,
linux-serial@vger.kernel.org, linux-omap@vger.kernel.org,
devicetree@vger.kernel.org, linux-arm-kernel@lists.infradead.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH 3/4] tty: omap-serial: use threaded interrupt handler
Date: Mon, 15 Sep 2014 13:31:56 -0400 [thread overview]
Message-ID: <5417228C.5070503@hurleysoftware.com> (raw)
In-Reply-To: <54170823.7040801@hurleysoftware.com>
On 09/15/2014 11:39 AM, Peter Hurley wrote:
> On 09/15/2014 10:00 AM, Frans Klaver wrote:
>> At 3.6Mbaud, with slightly over 2Mbit/s data coming in, we see 1600 uart
>> rx buffer overflows within 30 seconds. Threading the interrupt handling reduces
>> this to about 170 overflows in 10 minutes.
>
> Why is the threadirqs kernel boot option not sufficient?
> Or conversely, shouldn't this be selectable?
Also, do you see the same performance differential when you implement this
in the 8250 driver (that is, on top of Sebastian's omap->8250 conversion)?
> Regards,
> Peter Hurley
>
> PS - To overflow the 64 byte RX FIFO at those data rates means interrupt
> latency in excess of 250us?
>
>> In practice this therefore reduces the need for hardware flow control,
>> meaning the sending side doesn't have to buffer as much either.
>>
>> Signed-off-by: Frans Klaver <frans.klaver@xsens.com>
>> ---
>> drivers/tty/serial/omap-serial.c | 30 +++++++++++++++++++++++-------
>> 1 file changed, 23 insertions(+), 7 deletions(-)
>>
>> diff --git a/drivers/tty/serial/omap-serial.c b/drivers/tty/serial/omap-serial.c
>> index 7d3f557..398139a 100644
>> --- a/drivers/tty/serial/omap-serial.c
>> +++ b/drivers/tty/serial/omap-serial.c
>> @@ -575,6 +575,20 @@ static void serial_omap_rdi(struct uart_omap_port *up, unsigned int lsr)
>> }
>>
>> /**
>> + * serial_omap_fast_irq() - schedule interrupt handling
>> + */
>> +static irqreturn_t serial_omap_fast_irq(int irq, void *dev_id)
>> +{
>> + struct uart_omap_port *up = dev_id;
>> + unsigned int iir = serial_in(up, UART_IIR);
>> +
>> + if (iir & UART_IIR_NO_INT)
>> + return IRQ_NONE;
>> +
>> + return IRQ_WAKE_THREAD;
>> +}
>> +
>> +/**
>> * serial_omap_irq() - This handles the interrupt from one port
>> * @irq: uart port irq number
>> * @dev_id: uart port info
>> @@ -584,7 +598,6 @@ static irqreturn_t serial_omap_irq(int irq, void *dev_id)
>> struct uart_omap_port *up = dev_id;
>> unsigned int iir, lsr;
>> unsigned int type;
>> - irqreturn_t ret = IRQ_NONE;
>> int max_count = 256;
>>
>> spin_lock(&up->port.lock);
>> @@ -595,7 +608,6 @@ static irqreturn_t serial_omap_irq(int irq, void *dev_id)
>> if (iir & UART_IIR_NO_INT)
>> break;
>>
>> - ret = IRQ_HANDLED;
>> lsr = serial_in(up, UART_LSR);
>>
>> /* extract IRQ type from IIR register */
>> @@ -634,7 +646,7 @@ static irqreturn_t serial_omap_irq(int irq, void *dev_id)
>> pm_runtime_put_autosuspend(up->dev);
>> up->port_activity = jiffies;
>>
>> - return ret;
>> + return IRQ_HANDLED;
>> }
>>
>> static unsigned int serial_omap_tx_empty(struct uart_port *port)
>> @@ -731,15 +743,19 @@ static int serial_omap_startup(struct uart_port *port)
>> /*
>> * Allocate the IRQ
>> */
>> - retval = request_irq(up->port.irq, serial_omap_irq, up->port.irqflags,
>> - up->name, up);
>> + retval = request_threaded_irq(up->port.irq, serial_omap_fast_irq,
>> + serial_omap_irq,
>> + IRQF_ONESHOT | up->port.irqflags,
>> + up->name, up);
>> if (retval)
>> return retval;
>>
>> /* Optional wake-up IRQ */
>> if (up->wakeirq) {
>> - retval = request_irq(up->wakeirq, serial_omap_irq,
>> - up->port.irqflags, up->name, up);
>> + retval = request_threaded_irq(up->wakeirq, serial_omap_fast_irq,
>> + serial_omap_irq,
>> + IRQF_ONESHOT | up->port.irqflags,
>> + up->name, up);
>> if (retval) {
>> free_irq(up->port.irq, up);
>> return retval;
>>
next prev parent reply other threads:[~2014-09-15 17:31 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-09-15 14:00 [PATCH v3 0/4] omap-serial high-speed fixes/improvements Frans Klaver
2014-09-15 14:00 ` Frans Klaver
2014-09-15 14:00 ` Frans Klaver
2014-09-15 14:00 ` [PATCH 1/4] tty: omap-serial: pull out calculation from baud_is_mode16 Frans Klaver
2014-09-15 14:00 ` Frans Klaver
2014-09-15 14:00 ` Frans Klaver
2014-09-15 14:00 ` [PATCH 2/4] tty: omap-serial: prevent division by zero Frans Klaver
2014-09-15 14:00 ` Frans Klaver
2014-09-15 14:00 ` Frans Klaver
2014-09-15 14:00 ` [PATCH 3/4] tty: omap-serial: use threaded interrupt handler Frans Klaver
2014-09-15 14:00 ` Frans Klaver
2014-09-15 14:00 ` Frans Klaver
2014-09-15 15:39 ` Peter Hurley
2014-09-15 15:39 ` Peter Hurley
[not found] ` <54170823.7040801-WaGBZJeGNqdsbIuE7sb01tBPR1lH4CV8@public.gmane.org>
2014-09-15 17:31 ` Peter Hurley [this message]
2014-09-15 17:31 ` Peter Hurley
2014-09-15 17:31 ` Peter Hurley
2014-09-16 8:50 ` Frans Klaver
2014-09-16 8:50 ` Frans Klaver
2014-09-16 8:50 ` Frans Klaver
2014-09-17 12:01 ` Peter Hurley
2014-09-17 12:01 ` Peter Hurley
2014-09-17 12:13 ` Frans Klaver
2014-09-17 12:13 ` Frans Klaver
2014-09-17 12:13 ` Frans Klaver
2014-09-23 8:24 ` Frans Klaver
2014-09-23 8:24 ` Frans Klaver
2014-09-23 8:24 ` Frans Klaver
[not found] ` <20140923082445.GA16218-fflvphBLEC9V3v/g3MTWbjg1oUeLRpcR@public.gmane.org>
2014-09-23 17:17 ` Peter Hurley
2014-09-23 17:17 ` Peter Hurley
2014-09-23 17:17 ` Peter Hurley
[not found] ` <5421AB20.8040606-WaGBZJeGNqdsbIuE7sb01tBPR1lH4CV8@public.gmane.org>
2014-09-23 18:11 ` Frans Klaver
2014-09-23 18:11 ` Frans Klaver
2014-09-23 18:11 ` Frans Klaver
2014-09-23 18:38 ` Tony Lindgren
2014-09-23 18:38 ` Tony Lindgren
2014-09-23 20:02 ` Frans Klaver
2014-09-23 20:02 ` Frans Klaver
2014-09-15 14:00 ` [PATCH 4/4] tty: omap-serial: support setting of hardware flow control in dts Frans Klaver
2014-09-15 14:00 ` Frans Klaver
2014-09-15 14:00 ` Frans Klaver
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=5417228C.5070503@hurleysoftware.com \
--to=peter-wagbzjegnqdsbiue7sb01tbpr1lh4cv8@public.gmane.org \
--cc=alexey.pelykh-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org \
--cc=balbi-l0cyMroinI0@public.gmane.org \
--cc=devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=frans.klaver-MHHw4NDrbWwAvxtiuMwx3w@public.gmane.org \
--cc=galak-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org \
--cc=gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org \
--cc=ijc+devicetree-KcIKpvwj1kUDXYZnReoRVg@public.gmane.org \
--cc=jslaby-AlSwsSmVLrQ@public.gmane.org \
--cc=linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-omap-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-serial-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=mark.rutland-5wv7dgnIgG8@public.gmane.org \
--cc=pawel.moll-5wv7dgnIgG8@public.gmane.org \
--cc=robh+dt-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org \
--cc=tony-4v6yS6AI5VpBDgjK7y7TUQ@public.gmane.org \
/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
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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.