can: flexcan: fix irq flooding by clearing all interrupt sources

PATCH Applied Wolfgang Grandegger 2 participants 10 messages Dec 12, 2011 – Jan 31, 2012

Subsystems (3)

Networking Network drivers CAN network drivers

lore

Share & tools

lore

Wolfgang Grandegger

author

2011-12-12 15:09 UTC #0 lore raw

PATCH 1/1 can: flexcan: fix irq flooding by clearing all interrupt sources

As pointed out by Reuben Dowle and Lothar Waßmann, the TWRN_INT, RWRN_INT, BOFF_INT interrupt sources need to be cleared as well to avoid interrupt flooding, at least for the Flexcan on i.MX28 SOCs. Furthermore, the interrupts are only cleared, if really one of those interrupt sources are pending (which is not the case for rx and tx done).

CC Reuben Dowle <[email protected]>

CC Lothar Waßmann <[email protected]>

Signed-off-by Wolfgang Grandegger <[email protected]>

---
 drivers/net/can/flexcan.c |    7 ++++++-
 1 files changed, 6 insertions(+), 1 deletions(-)

drivers/net/can/flexcan.c +6 -1

index 165a4c7..111f154 100644
@@ -118,6 +118,9 @@
 	(FLEXCAN_ESR_TWRN_INT | FLEXCAN_ESR_RWRN_INT | FLEXCAN_ESR_BOFF_INT)
 #define FLEXCAN_ESR_ERR_ALL \
 	(FLEXCAN_ESR_ERR_BUS | FLEXCAN_ESR_ERR_STATE)
+#define FLEXCAN_ESR_ALL_INT \
+	(FLEXCAN_ESR_TWRN_INT | FLEXCAN_ESR_RWRN_INT | \
+	 FLEXCAN_ESR_BOFF_INT | FLEXCAN_ESR_ERR_INT)
 
 /* FLEXCAN interrupt flag register (IFLAG) bits */
 #define FLEXCAN_TX_BUF_ID		8

@@ -577,7 +580,9 @@ static irqreturn_t flexcan_irq(int irq, void *dev_id)
 
 	reg_iflag1 = flexcan_read(&regs->iflag1);
 	reg_esr = flexcan_read(&regs->esr);
-	flexcan_write(FLEXCAN_ESR_ERR_INT, &regs->esr);	/* ACK err IRQ */
+	/* ACK all bus error and state change IRQ sources */
+	if (reg_esr & FLEXCAN_ESR_ALL_INT)
+		flexcan_write(reg_esr & FLEXCAN_ESR_ALL_INT, &regs->esr);
 
 	/*
 	 * schedule NAPI in case of:

Signature · 1 line

1.7.4.1

Marc Kleine-Budde

maintainer

2011-12-12 15:31 UTC #1 lore raw

Quoted 10 lines · Wolfgang Grandegger

As pointed out by Reuben Dowle and Lothar Waßmann, the TWRN_INT, RWRN_INT, BOFF_INT interrupt sources need to be cleared as well to avoid interrupt flooding, at least for the Flexcan on i.MX28 SOCs. Furthermore, the interrupts are only cleared, if really one of those interrupt sources are pending (which is not the case for rx and tx done).
CC Reuben Dowle <[email protected]>
CC Lothar Waßmann <[email protected]>
Signed-off-by Wolfgang Grandegger <[email protected]>

Have you tested on mx25/mx35, does it have any negative side effects? My schedule is full until Friday, sorry cannot test here.

Marc

Quoted 29 lines

---
 drivers/net/can/flexcan.c |    7 ++++++-
 1 files changed, 6 insertions(+), 1 deletions(-)

drivers/net/can/flexcan.c +6 -1

index 165a4c7..111f154 100644
@@ -118,6 +118,9 @@
 	(FLEXCAN_ESR_TWRN_INT | FLEXCAN_ESR_RWRN_INT | FLEXCAN_ESR_BOFF_INT)
 #define FLEXCAN_ESR_ERR_ALL \
 	(FLEXCAN_ESR_ERR_BUS | FLEXCAN_ESR_ERR_STATE)
+#define FLEXCAN_ESR_ALL_INT \
+	(FLEXCAN_ESR_TWRN_INT | FLEXCAN_ESR_RWRN_INT | \
+	 FLEXCAN_ESR_BOFF_INT | FLEXCAN_ESR_ERR_INT)
 
 /* FLEXCAN interrupt flag register (IFLAG) bits */
 #define FLEXCAN_TX_BUF_ID		8

@@ -577,7 +580,9 @@ static irqreturn_t flexcan_irq(int irq, void *dev_id)
 
 	reg_iflag1 = flexcan_read(&regs->iflag1);
 	reg_esr = flexcan_read(&regs->esr);
-	flexcan_write(FLEXCAN_ESR_ERR_INT, &regs->esr);	/* ACK err IRQ */
+	/* ACK all bus error and state change IRQ sources */
+	if (reg_esr & FLEXCAN_ESR_ALL_INT)
+		flexcan_write(reg_esr & FLEXCAN_ESR_ALL_INT, &regs->esr);
 
 	/*
 	 * schedule NAPI in case of:

Signature · 4 lines

Pengutronix e.K.                  | Marc Kleine-Budde           |
Industrial Linux Solutions        | Phone: +49-231-2826-924     |
Vertretung West/Dortmund          | Fax:   +49-5121-206917-5555 |
Amtsgericht Hildesheim, HRA 2686  | http://www.pengutronix.de   |

Wolfgang Grandegger

author

2011-12-12 15:44 UTC #2 lore raw

Quoted 8 lines · Marc Kleine-Budde

Quoted 7 lines · Wolfgang Grandegger
As pointed out by Reuben Dowle and Lothar Waßmann, the TWRN_INT, RWRN_INT, BOFF_INT interrupt sources need to be cleared as well to avoid interrupt flooding, at least for the Flexcan on i.MX28 SOCs. Furthermore, the interrupts are only cleared, if really one of those interrupt sources are pending (which is not the case for rx and tx done).

Quoted 6 lines

Quoted 3 lines
CC Reuben Dowle <[email protected]>
CC Lothar Waßmann <[email protected]>
Signed-off-by Wolfgang Grandegger <[email protected]>
Have you tested on mx25/mx35, does it have any negative side effects? My schedule is full until Friday, sorry cannot test here.

Not yet. But it's not critical. Only the pending interrupt flags are cleared. Maybe somebody else out there could do some testing... before I get hold of a MX35PDK board.

Wolfgang.

Wolfgang Grandegger

author

2011-12-13 12:13 UTC #3 lore raw

Quoted 13 lines · Wolfgang Grandegger

Quoted 12 lines · Marc Kleine-Budde
Quoted 10 lines · Wolfgang Grandegger
As pointed out by Reuben Dowle and Lothar Waßmann, the TWRN_INT, RWRN_INT, BOFF_INT interrupt sources need to be cleared as well to avoid interrupt flooding, at least for the Flexcan on i.MX28 SOCs. Furthermore, the interrupts are only cleared, if really one of those interrupt sources are pending (which is not the case for rx and tx done).
CC Reuben Dowle <[email protected]>
CC Lothar Waßmann <[email protected]>
Signed-off-by Wolfgang Grandegger <[email protected]>

Quoted 6 lines

Quoted 2 lines
Have you tested on mx25/mx35, does it have any negative side effects? My schedule is full until Friday, sorry cannot test here.
Not yet. But it's not critical. Only the pending interrupt flags are cleared. Maybe somebody else out there could do some testing... before I get hold of a MX35PDK board.

I got my MX35PDK board working and can confirm, that the patch works on a i.mx35 as well. My testing also confirms, that the ESR TWRN_INT, RWRN_INT, FLEXCAN_ESR_BOFF_INT do not function as documented. These flags do show up once, together with ERR_INT, and then, after clearing, never again. Obviously a bug in the Flexcan logic. From the feedback we can say, that only the i.MX28 does behave differently (==correctly). All other seem to work with the current code:

Flexcan on
- i.mx25
- i.mx35
- i.mx53
- P1010/P1020

Wolfgang.

Marc Kleine-Budde

maintainer

2011-12-13 12:53 UTC #4 lore raw

Quoted 29 lines · Wolfgang Grandegger

Quoted 13 lines · Wolfgang Grandegger
Quoted 12 lines · Marc Kleine-Budde
Quoted 10 lines · Wolfgang Grandegger
As pointed out by Reuben Dowle and Lothar Waßmann, the TWRN_INT, RWRN_INT, BOFF_INT interrupt sources need to be cleared as well to avoid interrupt flooding, at least for the Flexcan on i.MX28 SOCs. Furthermore, the interrupts are only cleared, if really one of those interrupt sources are pending (which is not the case for rx and tx done).
CC Reuben Dowle <[email protected]>
CC Lothar Waßmann <[email protected]>
Signed-off-by Wolfgang Grandegger <[email protected]>
Quoted 6 lines
Quoted 2 lines
Have you tested on mx25/mx35, does it have any negative side effects? My schedule is full until Friday, sorry cannot test here.
Not yet. But it's not critical. Only the pending interrupt flags are cleared. Maybe somebody else out there could do some testing... before I get hold of a MX35PDK board.
I got my MX35PDK board working and can confirm, that the patch works on a i.mx35 as well. My testing also confirms, that the ESR TWRN_INT, RWRN_INT, FLEXCAN_ESR_BOFF_INT do not function as documented. These flags do show up once, together with ERR_INT, and then, after clearing, never again. Obviously a bug in the Flexcan logic. From the feedback we can say, that only the i.MX28 does behave differently (==correctly). All other seem to work with the current code:

Quoted 5 lines

Flexcan on
- i.mx25
- i.mx35
- i.mx53
- P1010/P1020

I'm adding the patch to linux-can. I think this is a stable candidate for v2.6.39 and newer (mx28 suport was added in 2.6.39).

Marc

Signature · 4 lines

Pengutronix e.K.                  | Marc Kleine-Budde           |
Industrial Linux Solutions        | Phone: +49-231-2826-924     |
Vertretung West/Dortmund          | Fax:   +49-5121-206917-5555 |
Amtsgericht Hildesheim, HRA 2686  | http://www.pengutronix.de   |

Wolfgang Grandegger

author

2011-12-13 12:53 UTC #5 lore raw

In reply to Wolfgang Grandegger

Quoted 29 lines · Wolfgang Grandegger

Quoted 13 lines · Wolfgang Grandegger
Quoted 12 lines · Marc Kleine-Budde
Quoted 10 lines · Wolfgang Grandegger
As pointed out by Reuben Dowle and Lothar Waßmann, the TWRN_INT, RWRN_INT, BOFF_INT interrupt sources need to be cleared as well to avoid interrupt flooding, at least for the Flexcan on i.MX28 SOCs. Furthermore, the interrupts are only cleared, if really one of those interrupt sources are pending (which is not the case for rx and tx done).
CC Reuben Dowle <[email protected]>
CC Lothar Waßmann <[email protected]>
Signed-off-by Wolfgang Grandegger <[email protected]>
Quoted 6 lines
Quoted 2 lines
Have you tested on mx25/mx35, does it have any negative side effects? My schedule is full until Friday, sorry cannot test here.
Not yet. But it's not critical. Only the pending interrupt flags are cleared. Maybe somebody else out there could do some testing... before I get hold of a MX35PDK board.
I got my MX35PDK board working and can confirm, that the patch works on a i.mx35 as well. My testing also confirms, that the ESR TWRN_INT, RWRN_INT, FLEXCAN_ESR_BOFF_INT do not function as documented. These flags do show up once, together with ERR_INT, and then, after clearing, never again. Obviously a bug in the Flexcan logic. From the feedback we can say, that only the i.MX28 does behave differently (==correctly). All other seem to work with the current code:

Quoted 5 lines

Flexcan on
- i.mx25
- i.mx35
- i.mx53
- P1010/P1020

But unfortunately, state change reporting looks different with this patch :)

Wolfgang.

Marc Kleine-Budde

maintainer

2011-12-13 12:59 UTC #6 lore raw

Quoted 8 lines · Wolfgang Grandegger

Quoted 8 lines
I got my MX35PDK board working and can confirm, that the patch works on a i.mx35 as well. My testing also confirms, that the ESR TWRN_INT, RWRN_INT, FLEXCAN_ESR_BOFF_INT do not function as documented. These flags do show up once, together with ERR_INT, and then, after clearing, never again. Obviously a bug in the Flexcan logic. From the feedback we can say, that only the i.MX28 does behave differently (==correctly). All other seem to work with the current code:

Quoted 7 lines

Quoted 5 lines
Flexcan on
- i.mx25
- i.mx35
- i.mx53
- P1010/P1020
But unfortunately, state change reporting looks different with this patch :)

Hmm - so not schedule for stable. What about your buf-off-handling, will this change the reporting again?

Marc

Signature · 4 lines

Pengutronix e.K.                  | Marc Kleine-Budde           |
Industrial Linux Solutions        | Phone: +49-231-2826-924     |
Vertretung West/Dortmund          | Fax:   +49-5121-206917-5555 |
Amtsgericht Hildesheim, HRA 2686  | http://www.pengutronix.de   |

Wolfgang Grandegger

author

2011-12-13 16:22 UTC #7 lore raw

Quoted 15 lines · Marc Kleine-Budde

Quoted 14 lines · Wolfgang Grandegger
Quoted 13 lines
I got my MX35PDK board working and can confirm, that the patch works on a i.mx35 as well. My testing also confirms, that the ESR TWRN_INT, RWRN_INT, FLEXCAN_ESR_BOFF_INT do not function as documented. These flags do show up once, together with ERR_INT, and then, after clearing, never again. Obviously a bug in the Flexcan logic. From the feedback we can say, that only the i.MX28 does behave differently (==correctly). All other seem to work with the current code:
Flexcan on
- i.mx25
- i.mx35
- i.mx53
- P1010/P1020

Quoted 4 lines

Quoted 1 line
But unfortunately, state change reporting looks different with this patch :)
Hmm - so not schedule for stable. What about your buf-off-handling, will this change the reporting again?

Well, as it is a serious problem on i.MX28, I would schedule this patch for stable as well. The error and state change reporting is bogus on the Flexcan anyhow. Without this patch, I get "active->warning->passive" if I send a message with cable disconnect (no ack). With patch just "active->warning". That's the same behaviour
as on the i.MX28, also with my new state and bus-off handling. See:

https://gitorious.org/~wgrandegger/linux-can/wg-linux-can-next/commit/bd3acb12dbb9551541d28ae8766c154d3cf6ed57

Wolfgang.

Wolfgang Grandegger

author

2011-12-14 13:21 UTC #8 lore raw

Hi Marc,

Quoted 21 lines · Wolfgang Grandegger

Quoted 15 lines · Marc Kleine-Budde
Quoted 14 lines · Wolfgang Grandegger
Quoted 13 lines
I got my MX35PDK board working and can confirm, that the patch works on a i.mx35 as well. My testing also confirms, that the ESR TWRN_INT, RWRN_INT, FLEXCAN_ESR_BOFF_INT do not function as documented. These flags do show up once, together with ERR_INT, and then, after clearing, never again. Obviously a bug in the Flexcan logic. From the feedback we can say, that only the i.MX28 does behave differently (==correctly). All other seem to work with the current code:
Flexcan on
- i.mx25
- i.mx35
- i.mx53
- P1010/P1020
Quoted 4 lines
Quoted 1 line
But unfortunately, state change reporting looks different with this patch :)
Hmm - so not schedule for stable. What about your buf-off-handling, will this change the reporting again?

Quoted 6 lines

Well, as it is a serious problem on i.MX28, I would schedule this patch for stable as well. The error and state change reporting is bogus on the Flexcan anyhow. Without this patch, I get "active->warning->passive" if I send a message with cable disconnect (no ack). With patch just "active->warning". That's the same behaviour
as on the i.MX28, also with my new state and bus-off handling. See:

I now understand the difference. I "flexcan_irq" we have:

	/*
	 * schedule NAPI in case of:
	 * - rx IRQ
	 * - state change IRQ
	 * - bus error IRQ and bus error reporting is activated
	 */
	if ((reg_iflag1 & FLEXCAN_IFLAG_RX_FIFO_AVAILABLE) ||
	    (reg_esr & FLEXCAN_ESR_ERR_STATE) ||
	    flexcan_has_and_handle_berr(priv, reg_esr)) {

Without this patch, "reg_esr & FLEXCAN_ESR_ERR_STATE" is *always* true because the RWRN_INT is never cleared. This means, *any* message is scheduled. As a nice side effect, this patch fixes this bug a well.

Only because any message is scheduled, the state change to error passive is recognized in flexcan_poll(). The state change to error passive is not signaled by an extra interrupt and is therefore only visible together with RX, TX done or bus-error events.

Wolfgang.

Marc Kleine-Budde

maintainer Applied

2012-01-31 21:59 UTC #9 lore raw

In reply to Wolfgang Grandegger

Quoted 10 lines · Wolfgang Grandegger

As pointed out by Reuben Dowle and Lothar Waßmann, the TWRN_INT, RWRN_INT, BOFF_INT interrupt sources need to be cleared as well to avoid interrupt flooding, at least for the Flexcan on i.MX28 SOCs. Furthermore, the interrupts are only cleared, if really one of those interrupt sources are pending (which is not the case for rx and tx done).
CC Reuben Dowle <[email protected]>
CC Lothar Waßmann <[email protected]>
Signed-off-by Wolfgang Grandegger <[email protected]>

Applied to can-next.
tnx Marc

Signature · 4 lines

Pengutronix e.K.                  | Marc Kleine-Budde           |
Industrial Linux Solutions        | Phone: +49-231-2826-924     |
Vertretung West/Dortmund          | Fax:   +49-5121-206917-5555 |
Amtsgericht Hildesheim, HRA 2686  | http://www.pengutronix.de   |