From: David Jander <david@protonic.nl>
To: Marc Kleine-Budde <mkl@pengutronix.de>
Cc: Wolfgang Grandegger <wg@grandegger.com>,
linux-can@vger.kernel.org,
Alexander Stein <alexander.stein@systec-electronic.com>
Subject: Re: [PATCH 11/15] can: rx-fifo: Add support for can state tracking and error polling
Date: Mon, 3 Nov 2014 13:51:35 +0100 [thread overview]
Message-ID: <20141103135135.10fd4c6f@archvile> (raw)
In-Reply-To: <545765D8.7090906@pengutronix.de>
On Mon, 03 Nov 2014 12:24:08 +0100
Marc Kleine-Budde <mkl@pengutronix.de> wrote:
> On 10/10/2014 05:46 PM, David Jander wrote:
> > The interrupt handler should store the error flags if needed and call
> > can_rx_fifo_irq_error() if there was an error flag set. This will trigger
> > a napi-poll that will call poll_can_state() and poll_bus_error()
> > callbacks. Those should handle can state machine and send error frames
> > as needed.
> >
> > Signed-off-by: David Jander <david@protonic.nl>
> > ---
> > drivers/net/can/dev.c | 23 ++++++++++++++++++++---
> > include/linux/can/dev.h | 9 +++++++++
> > 2 files changed, 29 insertions(+), 3 deletions(-)
> >
> > diff --git a/drivers/net/can/dev.c b/drivers/net/can/dev.c
> > index fc35d28..dac7579 100644
> > --- a/drivers/net/can/dev.c
> > +++ b/drivers/net/can/dev.c
> > @@ -352,6 +352,9 @@ static int can_rx_fifo_napi_poll(struct napi_struct
> > *napi, int quota) unsigned int tail;
> >
> > restart_poll:
> > + if (work_done < quota)
> > + work_done += fifo->poll_can_state(fifo);
> > +
>
> Do we need two callbacks in the poll-loop? I think one should be enough.
> The driver can call as many non-rx related functions as needed then.
> E.g. state change, bus error handling, tx-completion handling, etc...
Ok. I split it into two handlers, because I saw most drivers doing first
state-handling then message reception and last error handling... in this
particular order. If that order is not important, we can use one poll()
function for both state- and error-handling.
In that case... should the poll() handler be called _before_ or _after_
handling the mailboxes?
> What about adding a guard:
>
> if (fifo->poll && work_done < quota)
> work_done += fifo->poll(fifo);
>
> ...so that the poll callback can be NULL.
Ok, good idea.
> > /* handle mailboxes */
> > head = smp_load_acquire(&fifo->ring_head);
> > tail = fifo->ring_tail;
> > @@ -363,14 +366,19 @@ restart_poll:
> > smp_store_release(&fifo->ring_tail, tail);
> > }
> >
> > + if (work_done < quota)
> > + work_done += fifo->poll_bus_error(fifo);
> > +
> > if (work_done < quota) {
> > napi_complete(napi);
> >
> > /* Check if there was another interrupt */
> > head = smp_load_acquire(&fifo->ring_head);
> > - if ((CIRC_CNT(head, tail, fifo->ring_size) >= 1) &&
> > - napi_reschedule(&fifo->napi))
> > + if (((CIRC_CNT(head, tail, fifo->ring_size) >= 1) ||
> > + fifo->poll_errors) && napi_reschedule(&fifo->napi)) {
> > + fifo->poll_errors = false;
> > goto restart_poll;
> > + }
> > }
> >
> > return work_done;
> > @@ -393,7 +401,8 @@ int can_rx_fifo_add(struct net_device *dev, struct
> > can_rx_fifo *fifo) return -EINVAL;
> >
> > if (!fifo->mailbox_enable_mask || !fifo->mailbox_move_to_buffer ||
> > - !fifo->mailbox_enable)
> > + !fifo->mailbox_enable || !fifo->poll_bus_error ||
> > + !fifo->poll_can_state)
> > return -EINVAL;
> >
> > /* Make ring-buffer a sensible size that is a power of 2 */
> > @@ -412,6 +421,7 @@ int can_rx_fifo_add(struct net_device *dev, struct
> > can_rx_fifo *fifo) fifo->mask_low = can_rx_fifo_mask_low(fifo);
> > fifo->mask_high = can_rx_fifo_mask_high(fifo);
> > fifo->second_first = false;
> > + fifo->poll_errors = false;
> > fifo->active = fifo->mask_low | fifo->mask_high;
> > fifo->mailbox_enable_mask(fifo, fifo->active);
> >
> > @@ -503,6 +513,13 @@ int can_rx_fifo_irq_offload(struct can_rx_fifo *fifo)
> > }
> > EXPORT_SYMBOL_GPL(can_rx_fifo_irq_offload);
> >
> > +void can_rx_fifo_irq_error(struct can_rx_fifo *fifo)
> > +{
> > + fifo->poll_errors = true;
> > + can_rx_fifo_napi_schedule(fifo);
> > +}
> > +EXPORT_SYMBOL_GPL(can_rx_fifo_irq_error);
> > +
> > void can_rx_fifo_napi_enable(struct can_rx_fifo *fifo)
> > {
> > napi_enable(&fifo->napi);
> > diff --git a/include/linux/can/dev.h b/include/linux/can/dev.h
> > index c49c37b..e1ed6d4 100644
> > --- a/include/linux/can/dev.h
> > +++ b/include/linux/can/dev.h
> > @@ -75,12 +75,15 @@ struct can_rx_fifo {
> > void (*mailbox_enable)(struct can_rx_fifo *rx_fifo, unsigned int
> > mb); unsigned int (*mailbox_move_to_buffer)(struct can_rx_fifo *rx_fifo,
> > struct can_frame *frame, unsigned int mb);
> > + unsigned int (*poll_can_state)(struct can_rx_fifo *rx_fifo);
> > + unsigned int (*poll_bus_error)(struct can_rx_fifo *rx_fifo);
> >
> > u64 mask_low;
> > u64 mask_high;
> > u64 active;
> >
> > unsigned int second_first;
> > + bool poll_errors;
> >
> > bool inc;
> >
> > @@ -92,6 +95,11 @@ struct can_rx_fifo {
> > struct napi_struct napi;
> > };
> >
> > +static inline void can_rx_fifo_napi_schedule(struct can_rx_fifo *fifo)
> > +{
> > + napi_schedule(&fifo->napi);
> > +}
> > +
> > /*
> > * get_can_dlc(value) - helper macro to cast a given data length code
> > (dlc)
> > * to __u8 and ensure the dlc value to be max. 8 bytes.
> > @@ -135,6 +143,7 @@ u8 can_len2dlc(u8 len);
> >
> > int can_rx_fifo_add(struct net_device *dev, struct can_rx_fifo *fifo);
> > int can_rx_fifo_irq_offload(struct can_rx_fifo *fifo);
> > +void can_rx_fifo_irq_error(struct can_rx_fifo *fifo);
> > void can_rx_fifo_napi_enable(struct can_rx_fifo *fifo);
> > void can_rx_fifo_napi_disable(struct can_rx_fifo *fifo);
> > void can_rx_fifo_reset(struct can_rx_fifo *fifo);
> >
>
> Marc
>
Best regards,
--
David Jander
Protonic Holland.
next prev parent reply other threads:[~2014-11-03 12:51 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-10-10 15:46 [RFC PATCH V3 00/15] CAN: Add rx-fifo support and port flexcan to it David Jander
2014-10-10 15:46 ` [PATCH 01/15] can: flexcan: add documentation about mailbox organizaiton David Jander
2014-10-10 15:46 ` [PATCH 02/15] can: flexcan: rename crl2 -> ctrl2 David Jander
2014-10-10 15:46 ` [PATCH 03/15] can: flexcan: replace open coded mailbox code by proper defines David Jander
2014-10-10 15:46 ` [PATCH 04/15] can: flexcan: Re-write receive path to use MB queue instead of FIFO David Jander
2014-10-10 15:46 ` [PATCH 05/15] can: dev: add preliminary rx-fifo David Jander
2014-10-10 15:46 ` [PATCH 06/15] can: rx-fifo: Increase MB size limit from 32 to 64 David Jander
2014-10-19 21:25 ` Marc Kleine-Budde
2014-10-20 6:14 ` David Jander
2014-10-10 15:46 ` [PATCH 07/15] can: rx-fifo: Change to do controller off-load in interrupt and NAPI poll David Jander
2014-10-19 22:09 ` Marc Kleine-Budde
2014-10-20 7:06 ` David Jander
2014-11-03 11:10 ` Marc Kleine-Budde
2014-11-03 12:44 ` David Jander
2014-10-10 15:46 ` [PATCH 08/15] can: rx-fifo: fix long lines David Jander
2014-10-19 21:18 ` Marc Kleine-Budde
2014-10-20 7:09 ` David Jander
2014-10-10 15:46 ` [PATCH 09/15] can: rx-fifo: Add can_rx_fifo_reset() function David Jander
2014-11-03 11:16 ` Marc Kleine-Budde
2014-11-03 12:46 ` David Jander
2014-11-03 12:51 ` Marc Kleine-Budde
2014-10-10 15:46 ` [PATCH 10/15] can: rx-fifo: remove obsolete comment David Jander
2014-10-10 15:46 ` [PATCH 11/15] can: rx-fifo: Add support for can state tracking and error polling David Jander
2014-11-03 11:24 ` Marc Kleine-Budde
2014-11-03 12:51 ` David Jander [this message]
2014-11-03 12:58 ` Marc Kleine-Budde
2014-11-03 13:09 ` David Jander
2014-11-03 13:24 ` Marc Kleine-Budde
2014-11-05 17:16 ` David Jander
2014-11-06 10:20 ` Marc Kleine-Budde
2014-11-06 11:07 ` David Jander
2014-10-10 15:46 ` [PATCH 12/15] can: flexcan: Add support for RX-FIFO David Jander
2014-11-03 11:26 ` Marc Kleine-Budde
2014-11-03 12:55 ` David Jander
2014-11-03 13:34 ` Marc Kleine-Budde
2014-10-10 15:46 ` [PATCH 13/15] can: rx-fifo: Add support for simple irq offloading David Jander
2014-11-03 11:59 ` Marc Kleine-Budde
2014-10-10 15:46 ` [PATCH 14/15] can: flexcan: Add MB/Fifo specific column to comment table of IP versions David Jander
2014-10-10 15:47 ` [PATCH 15/15] can: flexcan: Re-enable RTR reception support for older flexcan IPs David Jander
2014-11-03 12:02 ` Marc Kleine-Budde
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=20141103135135.10fd4c6f@archvile \
--to=david@protonic.nl \
--cc=alexander.stein@systec-electronic.com \
--cc=linux-can@vger.kernel.org \
--cc=mkl@pengutronix.de \
--cc=wg@grandegger.com \
/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 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).