@@ -0,0 +1,24 @@+* Holt HI-311X stand-alone CAN controller device tree bindings++Required properties:+ - compatible: Should be one of the following:+ - "holt,hi3110" for HI-3110+ - reg: SPI chip select.+ - clocks: The clock feeding the CAN controller.+ - interrupt-parent: The parent interrupt controller.+ - interrupts: Should contain IRQ line for the CAN controller.++Optional properties:+ - vdd-supply: Regulator that powers the CAN controller.+ - xceiver-supply: Regulator that powers the CAN transceiver.++Example:+ can0: can@1 {+ compatible = "holt,hi3110";+ reg = <1>;+ clocks = <&clk32m>;+ interrupt-parent = <&gpio4>;+ interrupts = <13 IRQ_TYPE_EDGE_RISING>;+ vdd-supply = <®5v0>;+ xceiver-supply = <®5v0>;+ };
This patch adds support for the Holt HI-311x CAN controller. The HI311x
CAN controller is capable of transmitting and receiving standard data
frames, extended data frames and remote frames. The HI311x interfaces
with the host over SPI.
Datasheet: www.holtic.com/documents/371-hi-3110_v-rev-jpdf.do
Signed-off-by: Akshay Bhat <redacted>
---
v1 -> v2:
Address comments from Marc Kleine-Budde:
- use u8 instead of uint8_t
- alphabetically sort Makefile and Kconfig
- copy over copyright information from mcp251x
- use single space after each marco
- add missing HI3110_ namespace in defines
- remove magic number for IDE & SRR bits
- simplify logic to populate extended CAN ID
- remove unused parameters in hi3110_setup function
- remove redundant frame->can_dlc length check
- simplify error handling in hi3110_open function
Address comments from Julia Lawall:
- remove unnecessary semicolon after while loop in hi3110_can_ist
drivers/net/can/spi/Kconfig | 6 +
drivers/net/can/spi/Makefile | 1 +
drivers/net/can/spi/hi311x.c | 1069 ++++++++++++++++++++++++++++++++++++++++++
3 files changed, 1076 insertions(+)
create mode 100644 drivers/net/can/spi/hi311x.c
This patch adds support for the Holt HI-311x CAN controller. The HI311x
CAN controller is capable of transmitting and receiving standard data
frames, extended data frames and remote frames. The HI311x interfaces
with the host over SPI.
Datasheet: www.holtic.com/documents/371-hi-3110_v-rev-jpdf.do
Signed-off-by: Akshay Bhat <redacted>
---
Hi Marc,
Wanted to check if this patch can be included in the next kernel release
(4.12).
Thanks,
Akshay
From: Wolfgang Grandegger <hidden> Date: 2017-03-09 10:08:57
Hello Akshay,
unfortunately there are not many CAN controllers for the SPI bus. I just
know the MPC251x, which behaves badly (message losses) under Linux,
especially at hight bit-rates due to insufficient RX buffering. What is
your experience with that driver for the HI-311x?
Thanks,
Wolfgang.
Am 07.03.2017 um 16:31 schrieb Akshay Bhat:
On 01/17/2017 02:22 PM, Akshay Bhat wrote:
quoted
This patch adds support for the Holt HI-311x CAN controller. The HI311x
CAN controller is capable of transmitting and receiving standard data
frames, extended data frames and remote frames. The HI311x interfaces
with the host over SPI.
Datasheet: www.holtic.com/documents/371-hi-3110_v-rev-jpdf.do
Signed-off-by: Akshay Bhat <redacted>
---
Hi Marc,
Wanted to check if this patch can be included in the next kernel release
(4.12).
Thanks,
Akshay
--
To unsubscribe from this list: send the line "unsubscribe linux-can" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
On 03/09/2017 04:59 AM, Wolfgang Grandegger wrote:
Hello Akshay,
unfortunately there are not many CAN controllers for the SPI bus. I just
know the MPC251x, which behaves badly (message losses) under Linux,
especially at hight bit-rates due to insufficient RX buffering. What is
your experience with that driver for the HI-311x?
Hi Wolfgang,
Good question. I have not worked with MPC251x but the HI-311x performs
much better because HI-3110 features:
8 message FIFO (as opposed to 2 buffers on MPC2510)
20 MHz SPI interface (as opposed to 2.5 MHz on MPC2510)
As for the real world test results:
With RT patch applied to the kernel running on a i.MX6 Dual processor
(worst case interrupt latency of 50us as reported by cyclictest), there
are ZERO packet drops.
Tested with Kvaser Leaf sending 100 burst messages (back to back) every
40ms at a 1M CAN bit rate. 10 million messages were sent by the Kvaser
leaf and received successfully by the HI-311x driver.
Even without the RT patch, I was able to get the packet drop to zero but
this was by moving the CAN/SPI IRQ threads to CPU1 instead of CPU0.
Hence I feel the driver is a good candidate to be included in the Linux
kernel.
Below are detailed test results if you like:
https://goo.gl/VWgzp7
Ones of particular interest:
10M-msgs-1M-bitrate-100burst-40ms-interval-leaf.png
10M-msgs-1M-bitrate-100burst-40ms-interval-pwc.txt
Thanks,
Akshay
Am 07.03.2017 um 16:31 schrieb Akshay Bhat:
quoted
On 01/17/2017 02:22 PM, Akshay Bhat wrote:
quoted
This patch adds support for the Holt HI-311x CAN controller. The HI311x
CAN controller is capable of transmitting and receiving standard data
frames, extended data frames and remote frames. The HI311x interfaces
with the host over SPI.
Datasheet: www.holtic.com/documents/371-hi-3110_v-rev-jpdf.do
Signed-off-by: Akshay Bhat <redacted>
---
Hi Marc,
Wanted to check if this patch can be included in the next kernel release
(4.12).
Thanks,
Akshay
--
To unsubscribe from this list: send the line "unsubscribe linux-can" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
--
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
From: Wolfgang Grandegger <hidden> Date: 2017-03-09 14:45:33
Hello Akshay,
Am 09.03.2017 um 13:34 schrieb Akshay Bhat:
On 03/09/2017 04:59 AM, Wolfgang Grandegger wrote:
quoted
Hello Akshay,
unfortunately there are not many CAN controllers for the SPI bus. I just
know the MPC251x, which behaves badly (message losses) under Linux,
especially at hight bit-rates due to insufficient RX buffering. What is
your experience with that driver for the HI-311x?
Hi Wolfgang,
Good question. I have not worked with MPC251x but the HI-311x performs
much better because HI-3110 features:
8 message FIFO (as opposed to 2 buffers on MPC2510)
20 MHz SPI interface (as opposed to 2.5 MHz on MPC2510)
As for the real world test results:
With RT patch applied to the kernel running on a i.MX6 Dual processor
(worst case interrupt latency of 50us as reported by cyclictest), there
are ZERO packet drops.
Tested with Kvaser Leaf sending 100 burst messages (back to back) every
40ms at a 1M CAN bit rate. 10 million messages were sent by the Kvaser
leaf and received successfully by the HI-311x driver.
This corresponds to a bus load of approx. 50%, I think?
Even without the RT patch, I was able to get the packet drop to zero but
this was by moving the CAN/SPI IRQ threads to CPU1 instead of CPU0.
Vanilla Linux is more critical here due to higher latencies. With 2500
Messages per sec the RX FIFO (8 Messages) fills up within 3.2 ms... and
in a burst even quicker. That's already heavy load.
Wolfgang.
Hi Wolfgang,
On 03/09/2017 09:45 AM, Wolfgang Grandegger wrote:
Hello Akshay,
Am 09.03.2017 um 13:34 schrieb Akshay Bhat:
quoted
Hi Wolfgang,
Good question. I have not worked with MPC251x but the HI-311x performs
much better because HI-3110 features:
8 message FIFO (as opposed to 2 buffers on MPC2510)
20 MHz SPI interface (as opposed to 2.5 MHz on MPC2510)
As for the real world test results:
With RT patch applied to the kernel running on a i.MX6 Dual processor
(worst case interrupt latency of 50us as reported by cyclictest), there
are ZERO packet drops.
Tested with Kvaser Leaf sending 100 burst messages (back to back) every
40ms at a 1M CAN bit rate. 10 million messages were sent by the Kvaser
leaf and received successfully by the HI-311x driver.
This corresponds to a bus load of approx. 50%, I think?
The usecase I tested above was more like ~30% (data payload length was
randomized) but I did test with 50% bus load as well without any dropped
messages.
quoted
Even without the RT patch, I was able to get the packet drop to zero but
this was by moving the CAN/SPI IRQ threads to CPU1 instead of CPU0.
Vanilla Linux is more critical here due to higher latencies. With 2500
Messages per sec the RX FIFO (8 Messages) fills up within 3.2 ms... and
in a burst even quicker. That's already heavy load.
I agree with Vanilla Linux is more critical due to higher latencies,
however if the frequency governor is set to performance and CAN/SPI
threads are moved to another CPU (if that is an option) with SCHED_FIFO
setting, the driver is able to keep up even at high bit rates / bursts.
I understand SPI based CAN controllers are not ideal choice but for
users that do not have any other option, I feel this controller/driver
meets their need :)
Thanks,
Akshay
From: Wolfgang Grandegger <hidden> Date: 2017-03-09 17:37:50
Hello,
doing a quick review... I realized a few issues...
Am 17.01.2017 um 20:22 schrieb Akshay Bhat:
quoted hunk
This patch adds support for the Holt HI-311x CAN controller. The HI311x
CAN controller is capable of transmitting and receiving standard data
frames, extended data frames and remote frames. The HI311x interfaces
with the host over SPI.
Datasheet: www.holtic.com/documents/371-hi-3110_v-rev-jpdf.do
Signed-off-by: Akshay Bhat <redacted>
---
v1 -> v2:
Address comments from Marc Kleine-Budde:
- use u8 instead of uint8_t
- alphabetically sort Makefile and Kconfig
- copy over copyright information from mcp251x
- use single space after each marco
- add missing HI3110_ namespace in defines
- remove magic number for IDE & SRR bits
- simplify logic to populate extended CAN ID
- remove unused parameters in hi3110_setup function
- remove redundant frame->can_dlc length check
- simplify error handling in hi3110_open function
Address comments from Julia Lawall:
- remove unnecessary semicolon after while loop in hi3110_can_ist
drivers/net/can/spi/Kconfig | 6 +
drivers/net/can/spi/Makefile | 1 +
drivers/net/can/spi/hi311x.c | 1069 ++++++++++++++++++++++++++++++++++++++++++
3 files changed, 1076 insertions(+)
create mode 100644 drivers/net/can/spi/hi311x.c
The driver transfers the packets in sequence. Any chance to queue them?
At least there is a TX FIFO for 8 messages. That's bad for RT but would
increase the throughput.
+ priv->tx_skb = skb;
+ queue_work(priv->wq, &priv->tx_work);
+
+ return NETDEV_TX_OK;
+}
+
+static int hi3110_do_set_mode(struct net_device *net, enum can_mode mode)
+{
+ struct hi3110_priv *priv = netdev_priv(net);
+
+ switch (mode) {
+ case CAN_MODE_START:
+ hi3110_clean(net);
+ /* We have to delay work since SPI I/O may sleep */
+ priv->can.state = CAN_STATE_ERROR_ACTIVE;
+ priv->restart_tx = 1;
+ if (priv->can.restart_ms == 0)
+ priv->after_suspend = HI3110_AFTER_SUSPEND_RESTART;
+ queue_work(priv->wq, &priv->restart_work);
+ break;
+ default:
+ return -EOPNOTSUPP;
+ }
+
+ return 0;
+}
+
+static int hi3110_set_normal_mode(struct spi_device *spi)
+{
+ struct hi3110_priv *priv = spi_get_drvdata(spi);
+ u8 reg;
+
+ hi3110_write(spi, HI3110_WRITE_INTE, HI3110_INT_BUSERR |
+ HI3110_INT_RXFIFO | HI3110_INT_TXCPLT);
+
+ /* Enable TX */
+ hi3110_write(spi, HI3110_WRITE_CTRL1, HI3110_CTRL1_TXEN);
+
+ if (priv->can.ctrlmode & CAN_CTRLMODE_LOOPBACK) {
+ /* Put device into loopback mode */
+ hi3110_write(spi, HI3110_WRITE_CTRL0,
+ HI3110_CTRL0_LOOPBACK_MODE);
+ } else if (priv->can.ctrlmode & CAN_CTRLMODE_LISTENONLY) {
+ /* Put device into listen-only mode */
+ hi3110_write(spi, HI3110_WRITE_CTRL0,
+ HI3110_CTRL0_MONITOR_MODE);
+ } else {
+ /* Put device into normal mode */
+ hi3110_write(spi, HI3110_WRITE_CTRL0,
+ HI3110_CTRL0_NORMAL_MODE);
"mode = x" and just one write is more compact.
+
+ /* Wait for the device to enter normal mode */
+ mdelay(HI3110_OST_DELAY_MS);
+ reg = hi3110_read(spi, HI3110_READ_CTRL0);
+ if ((reg & HI3110_CTRL0_MODE_MASK) != HI3110_CTRL0_NORMAL_MODE)
+ return -EBUSY;
Is this not necesary for listen or loopbcak only mode?
And what about the ACK and CRC error defines at the beginning?
It's also comon to use netdev_dbg() on error interrupts.
+ }
Bus error reporting can flood the system with interrupts. Any chance to
implement CAN_CTRLMODE_BERR_REPORTING. I think the bus error interrupt
can be enabled/disabled.
+ /* Update can state statistics */
+ switch (priv->can.state) {
+ case CAN_STATE_ERROR_ACTIVE:
+ if (new_state >= CAN_STATE_ERROR_WARNING &&
+ new_state <= CAN_STATE_BUS_OFF)
+ priv->can.can_stats.error_warning++;
+ /* fallthrough */
+ case CAN_STATE_ERROR_WARNING:
+ if (new_state >= CAN_STATE_ERROR_PASSIVE &&
+ new_state <= CAN_STATE_BUS_OFF)
+ priv->can.can_stats.error_passive++;
+ break;
+ default:
+ break;
+ }
+ priv->can.state = new_state;
+
+ if (intf & HI3110_INT_BUSERR) {
+ /* Note: HI3110 Does report overflow errors */
+ hi3110_error_skb(net, can_id, data1, data2);
+ }
Usually the bus error counts are filled in the error message frame as
well. It the counts are available, I would also be nice to have the
"do_get_berr_counter" callback as well.
A few other things to check:
Run "cangen" and monitor the message with "candump -e any,0:0,#FFFFFFF".
Then 1) disconnect the cable or 2) short-circuit CAN low and high at the
connector. You should see error messages. After reconnection or removing
the short-circuit (and bus-off recovery) the state should go back to
"active".
Run "canfdtest" to check for out-of-order messages. I do not expect
problems with your driver, though.
Thanks,
Wolfgang.
The driver transfers the packets in sequence. Any chance to queue them?
At least there is a TX FIFO for 8 messages. That's bad for RT but would
increase the throughput.
I initially did not use the TX FIFO for the reason you mentioned above.
Queuing should be possible but since it requires lot more additional
logic, I can work on it a later time.
quoted
+
+ if (priv->can.ctrlmode & CAN_CTRLMODE_LOOPBACK) {
+ /* Put device into loopback mode */
+ hi3110_write(spi, HI3110_WRITE_CTRL0,
+ HI3110_CTRL0_LOOPBACK_MODE);
+ } else if (priv->can.ctrlmode & CAN_CTRLMODE_LISTENONLY) {
+ /* Put device into listen-only mode */
+ hi3110_write(spi, HI3110_WRITE_CTRL0,
+ HI3110_CTRL0_MONITOR_MODE);
+ } else {
+ /* Put device into normal mode */
+ hi3110_write(spi, HI3110_WRITE_CTRL0,
+ HI3110_CTRL0_NORMAL_MODE);
"mode = x" and just one write is more compact.
Fixed in v3 patch.
quoted
+
+ /* Wait for the device to enter normal mode */
+ mdelay(HI3110_OST_DELAY_MS);
+ reg = hi3110_read(spi, HI3110_READ_CTRL0);
+ if ((reg & HI3110_CTRL0_MODE_MASK) != HI3110_CTRL0_NORMAL_MODE)
+ return -EBUSY;
Is this not necesary for listen or loopbcak only mode?
And what about the ACK and CRC error defines at the beginning?
It's also comon to use netdev_dbg() on error interrupts.
Good catch, I missed it. Fixed in v3 patch.
quoted
+ }
Bus error reporting can flood the system with interrupts. Any chance to
implement CAN_CTRLMODE_BERR_REPORTING. I think the bus error interrupt
can be enabled/disabled.
Thanks, was not aware of this feature. Added it in v3 patch.
quoted
+ /* Update can state statistics */
+ switch (priv->can.state) {
+ case CAN_STATE_ERROR_ACTIVE:
+ if (new_state >= CAN_STATE_ERROR_WARNING &&
+ new_state <= CAN_STATE_BUS_OFF)
+ priv->can.can_stats.error_warning++;
+ /* fallthrough */
+ case CAN_STATE_ERROR_WARNING:
+ if (new_state >= CAN_STATE_ERROR_PASSIVE &&
+ new_state <= CAN_STATE_BUS_OFF)
+ priv->can.can_stats.error_passive++;
+ break;
+ default:
+ break;
+ }
+ priv->can.state = new_state;
+
+ if (intf & HI3110_INT_BUSERR) {
+ /* Note: HI3110 Does report overflow errors */
+ hi3110_error_skb(net, can_id, data1, data2);
+ }
Usually the bus error counts are filled in the error message frame as
well. It the counts are available, I would also be nice to have the
"do_get_berr_counter" callback as well.
Added it in v3 patch.
quoted
+static int hi3110_open(struct net_device *net)
+{
+ struct hi3110_priv *priv = netdev_priv(net);
+ struct spi_device *spi = priv->spi;
+ unsigned long flags = IRQF_ONESHOT | IRQF_TRIGGER_RISING;
+ int ret;
+
+ ret = open_candev(net);
+ if (ret) {
+ dev_err(&spi->dev, "unable to set initial baudrate!\n");
open_candev() does already print an error message.
Fixed in v3 patch.
A few other things to check:
Run "cangen" and monitor the message with "candump -e any,0:0,#FFFFFFF".
Then 1) disconnect the cable or 2) short-circuit CAN low and high at the
connector. You should see error messages. After reconnection or removing
the short-circuit (and bus-off recovery) the state should go back to
"active".
With the above sequence, candump reports "ERRORFRAME" with
protocol-violation{{}{acknowledge-slot}}, bus-error. On re-connecting
the cable the can state goes back to ACTIVE and I see the messages that
were in the queue being sent.
Run "canfdtest" to check for out-of-order messages. I do not expect
problems with your driver, though.
Ran canfdtest for 100k loop (@1M can bit rate), did not see any issues.
Thanks for your valuable suggests and tests :)
Akshay.
--
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
From: Wolfgang Grandegger <hidden> Date: 2017-03-14 12:11:21
Hallo Akshay,
Am 13.03.2017 um 16:38 schrieb Akshay Bhat:
Hi Wolfgang,
On 03/09/2017 12:36 PM, Wolfgang Grandegger wrote:
quoted
Hello,
doing a quick review... I realized a few issues...
Am 17.01.2017 um 20:22 schrieb Akshay Bhat:
... snip ...
quoted
A few other things to check:
Run "cangen" and monitor the message with "candump -e any,0:0,#FFFFFFF".
Then 1) disconnect the cable or 2) short-circuit CAN low and high at the
connector. You should see error messages. After reconnection or removing
the short-circuit (and bus-off recovery) the state should go back to
"active".
With the above sequence, candump reports "ERRORFRAME" with
protocol-violation{{}{acknowledge-slot}}, bus-error. On re-connecting
the cable the can state goes back to ACTIVE and I see the messages that
were in the queue being sent.
Do you get the ACK error also with berr-reporting off? Would be nice if
you could show a candump log here.
Also, any error message should show the bus error counts in data[7,8]:
http://lxr.free-electrons.com/source/drivers/net/can/sja1000/sja1000.c#L408
And please check bus-off as well (short-circuiting CAN low and high).
Wolfgang.
--
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
Hi Wolfgang,
On 03/14/2017 08:11 AM, Wolfgang Grandegger wrote:
... snip ...
quoted
quoted
A few other things to check:
Run "cangen" and monitor the message with "candump -e any,0:0,#FFFFFFF".
Then 1) disconnect the cable or 2) short-circuit CAN low and high at the
connector. You should see error messages. After reconnection or removing
the short-circuit (and bus-off recovery) the state should go back to
"active".
With the above sequence, candump reports "ERRORFRAME" with
protocol-violation{{}{acknowledge-slot}}, bus-error. On re-connecting
the cable the can state goes back to ACTIVE and I see the messages that
were in the queue being sent.
Do you get the ACK error also with berr-reporting off? Would be nice if
you could show a candump log here.
Below is a log for disconnecting and re-connecting CAN cable scenario:
(Note this is on a 4.1.18 kernel with RT patch)
root@imx6qrom5420b1:~# ip link set can0 up type can bitrate 1000000
berr-reporting on
root@imx6qrom5420b1:~# candump -e any,0:0,#FFFFFFF &
[1] 768
root@imx6qrom5420b1:~# cangen can0
can0 21C [8] 35 98 C0 7A 95 03 E6 2A
can0 6E6 [1] F2
can0 5C7 [2] 42 50
can0 57C [8] 83 7A E4 0C 03 8B 90 45
can0 55C [8] B9 74 87 52 D8 F4 64 04
can0 014 [8] 28 CB 96 57 3B 80 67 4F
can0 6AF [1] 35
can0 51E [8] B6 C8 6C 1D 3A 87 ED 2E
can0 527 [8] D0 8A D3 59 0E 34 40 78
can0 30C [2] 6A 12
can0 145 [8] CB 6E FF 55 C1 BE C3 22
can0 5A5 [8] C4 49 54 68 02 63 F9 35
can0 0BA [8] DA 57 5E 3A CE 88 20 1C
can0 516 [2] 09 09
can0 743 [8] 7C 4D 25 47 61 4C 56 3D
can0 31D [2] 9C D3
can0 71E [8] 53 7C 97 2A 2A F2 9F 56
can0 52E [8] FE DA 2D 51 73 96 DF 79
/////disconnect cable
can0 20000088 [8] 00 00 00 19 00 00 28 00 ERRORFRAME
protocol-violation{{}{acknowledge-slot}}
bus-error
error-counter-tx-rx{{40}{0}}
can0 20000088 [8] 00 00 00 19 00 00 58 00 ERRORFRAME
protocol-violation{{}{acknowledge-slot}}
bus-error
error-counter-tx-rx{{88}{0}}
can0 20000088 [8] 00 00 00 19 00 00 80 00 ERRORFRAME
protocol-violation{{}{acknowledge-slot}}
bus-error
error-counter-tx-rx{{128}{0}}
can0 2000008C [8] 00 20 00 19 00 00 80 00 ERRORFRAME
controller-problem{tx-error-passive}
protocol-violation{{}{acknowledge-slot}}
bus-error
error-counter-tx-rx{{128}{0}}
write: No buffer space available
root@imx6qrom5420b1:~# ip -s -d link show can0
4: can0: <NOARP,UP,LOWER_UP,ECHO> mtu 16 qdisc pfifo_fast state UNKNOWN
mode DEFAULT group default qlen 10
link/can promiscuity 0
can <BERR-REPORTING> state ERROR-PASSIVE (berr-counter tx 128 rx 0)
restart-ms 0
bitrate 1000000 sample-point 0.750
tq 62 prop-seg 5 phase-seg1 6 phase-seg2 4 sjw 1
hi3110: tseg1 2..16 tseg2 2..8 sjw 1..4 brp 1..64 brp-inc 1
clock 16000000
re-started bus-errors arbit-lost error-warn error-pass bus-off
0 6 0 1 1 0
RX: bytes packets errors dropped overrun mcast
0 0 6 0 0 0
TX: bytes packets errors dropped carrier collsns
106 18 0 0 0 0
root@imx6qrom5420b1:~#
/////re-connect cable
can0 169 [8] 35 55 A3 1C 0F 47 2E 5B
can0 318 [8] 11 AA 27 11 D2 1B CE 34
can0 577 [8] A0 A4 EE 50 8D A2 E1 3E
can0 4ED [8] 52 96 17 7E 31 FC 7D 7C
can0 2E7 [8] 92 48 D4 39 05 1E 9F 50
can0 200 [8] 4A 66 F6 02 1E 71 8E 26
can0 29A [8] 49 63 2E 7D C9 77 85 7A
can0 15A [7] 3C 0E 65 74 C3 62 80
can0 011 [1] D2
can0 26B [3] FC D6 68
can0 5CE [8] 6F 02 B5 14 BC 7A D7 02
root@imx6qrom5420b1:~# ip -s -d link show can0
4: can0: <NOARP,UP,LOWER_UP,ECHO> mtu 16 qdisc pfifo_fast state UNKNOWN
mode DEFAULT group default qlen 10
link/can promiscuity 0
can <BERR-REPORTING> state ERROR-ACTIVE (berr-counter tx 117 rx 0)
restart-ms 0
bitrate 1000000 sample-point 0.750
tq 62 prop-seg 5 phase-seg1 6 phase-seg2 4 sjw 1
hi3110: tseg1 2..16 tseg2 2..8 sjw 1..4 brp 1..64 brp-inc 1
clock 16000000
re-started bus-errors arbit-lost error-warn error-pass bus-off
0 7 0 1 1 0
RX: bytes packets errors dropped overrun mcast
0 0 7 0 0 0
TX: bytes packets errors dropped carrier collsns
181 29 0 0 0 0
//Reboot the board and test with bus error reporting off
root@imx6qrom5420b1:~# ip link set can0 up type can bitrate 1000000
berr-reporting off
root@imx6qrom5420b1:~# candump -e any,0:0,#FFFFFFF &
[1] 782
root@imx6qrom5420b1:~# cangen can0
can0 1FA [3] C9 FE C2
can0 3E2 [5] 85 37 03 5B 6F
can0 289 [8] A4 F6 BF 4A 3F 70 65 1B
can0 12D [8] B2 72 10 33 AB B4 68 64
can0 054 [2] 01 D7
can0 4A6 [8] 29 7D 76 56 CA C1 60 00
can0 768 [8] 97 3D 92 08 61 C1 D9 03
can0 098 [6] A4 A8 5A 60 92 1A
can0 3C9 [8] 71 78 0D 25 AB 27 8B 51
/////disconnect cable
write: No buffer space available
root@imx6qrom5420b1:~# ip -s -d link show can0
4: can0: <NOARP,UP,LOWER_UP,ECHO> mtu 16 qdisc pfifo_fast state UNKNOWN
mode DEFAULT group default qlen 10
link/can promiscuity 0
can state ERROR-ACTIVE (berr-counter tx 128 rx 0) restart-ms 0
bitrate 1000000 sample-point 0.750
tq 62 prop-seg 5 phase-seg1 6 phase-seg2 4 sjw 1
hi3110: tseg1 2..16 tseg2 2..8 sjw 1..4 brp 1..64 brp-inc 1
clock 16000000
re-started bus-errors arbit-lost error-warn error-pass bus-off
0 0 0 0 0 0
RX: bytes packets errors dropped overrun mcast
0 0 0 0 0 0
TX: bytes packets errors dropped carrier collsns
56 9 0 0 0 0
root@imx6qrom5420b1:~#
/////re-connect cable
can0 20000088 [8] 00 00 00 19 00 00 7F 00 ERRORFRAME
protocol-violation{{}{acknowledge-slot}}
bus-error
error-counter-tx-rx{{127}{0}}
can0 553 [6] 1A E4 60 6B DC 07
can0 7E3 [8] 1C 78 95 6E 10 81 AA 40
can0 20C [8] BB 35 13 25 60 0A 56 57
can0 1D0 [8] 48 4A 39 64 76 E6 57 08
can0 43A [1] 40
can0 2CF [7] 03 45 5E 0F 67 33 4C
can0 1CD [8] F9 4D AB 1D 96 A5 67 0E
can0 515 [8] 41 CD F2 5F 68 92 43 16
can0 661 [8] 45 9A 73 69 45 EE 8B 42
can0 41B [1] 55
can0 52F [1] 87
root@imx6qrom5420b1:~# ip -s -d link show can0
4: can0: <NOARP,UP,LOWER_UP,ECHO> mtu 16 qdisc pfifo_fast state UNKNOWN
mode DEFAULT group default qlen 10
link/can promiscuity 0
can state ERROR-ACTIVE (berr-counter tx 117 rx 0) restart-ms 0
bitrate 1000000 sample-point 0.750
tq 62 prop-seg 5 phase-seg1 6 phase-seg2 4 sjw 1
hi3110: tseg1 2..16 tseg2 2..8 sjw 1..4 brp 1..64 brp-inc 1
clock 16000000
re-started bus-errors arbit-lost error-warn error-pass bus-off
0 1 0 0 0 0
RX: bytes packets errors dropped overrun mcast
0 0 1 0 0 0
TX: bytes packets errors dropped carrier collsns
120 20 0 0 0 0
From: Wolfgang Grandegger <hidden> Date: 2017-03-14 18:08:37
Hello Akshay,
Am 14.03.2017 um 17:20 schrieb Akshay Bhat:
Hi Wolfgang,
On 03/14/2017 08:11 AM, Wolfgang Grandegger wrote:
quoted
... snip ...
quoted
quoted
A few other things to check:
Run "cangen" and monitor the message with "candump -e any,0:0,#FFFFFFF".
Then 1) disconnect the cable or 2) short-circuit CAN low and high at the
connector. You should see error messages. After reconnection or removing
the short-circuit (and bus-off recovery) the state should go back to
"active".
With the above sequence, candump reports "ERRORFRAME" with
protocol-violation{{}{acknowledge-slot}}, bus-error. On re-connecting
the cable the can state goes back to ACTIVE and I see the messages that
were in the queue being sent.
Do you get the ACK error also with berr-reporting off? Would be nice if
you could show a candump log here.
Below is a log for disconnecting and re-connecting CAN cable scenario:
(Note this is on a 4.1.18 kernel with RT patch)
root@imx6qrom5420b1:~# ip link set can0 up type can bitrate 1000000
berr-reporting on
root@imx6qrom5420b1:~# candump -e any,0:0,#FFFFFFF &
Here "tx-error-passiv" is packed with a bus error. What I'm looking for
are state change messages similar to:
can0 20000204 [8] 00 08 00 00 00 00 60 00 ERRORFRAME
controller-problem{tx-error-warning}
state-change{tx-error-warning}
error-counter-tx-rx{{96}{0}}
can0 20000204 [8] 00 30 00 00 00 00 80 00 ERRORFRAME
controller-problem{tx-error-passive}
state-change{tx-error-passive}
error-counter-tx-rx{{128}{0}
They should always come, even with "berr-reporting off".
write: No buffer space available
root@imx6qrom5420b1:~# ip -s -d link show can0
4: can0: <NOARP,UP,LOWER_UP,ECHO> mtu 16 qdisc pfifo_fast state UNKNOWN
mode DEFAULT group default qlen 10
link/can promiscuity 0
can <BERR-REPORTING> state ERROR-PASSIVE (berr-counter tx 128 rx 0)
restart-ms 0
bitrate 1000000 sample-point 0.750
tq 62 prop-seg 5 phase-seg1 6 phase-seg2 4 sjw 1
hi3110: tseg1 2..16 tseg2 2..8 sjw 1..4 brp 1..64 brp-inc 1
clock 16000000
re-started bus-errors arbit-lost error-warn error-pass bus-off
0 6 0 1 1 0
The error warning and passive counter increased , though. Also the bus
error should come in at a rather hight rate. Looking to the code, maybe
you need to test STATF to check for state changes (and not ERR).
After some more messages there should be also:
can0 20000200 [8] 00 40 00 00 00 00 5F 00 ERRORFRAME
state-change{back-to-error-active}
error-counter-tx-rx{{95}{0}}
For each message sent, the error counter decreases by 8.
root@imx6qrom5420b1:~# ip -s -d link show can0
4: can0: <NOARP,UP,LOWER_UP,ECHO> mtu 16 qdisc pfifo_fast state UNKNOWN
mode DEFAULT group default qlen 10
link/can promiscuity 0
can state ERROR-ACTIVE (berr-counter tx 117 rx 0) restart-ms 0
bitrate 1000000 sample-point 0.750
tq 62 prop-seg 5 phase-seg1 6 phase-seg2 4 sjw 1
hi3110: tseg1 2..16 tseg2 2..8 sjw 1..4 brp 1..64 brp-inc 1
clock 16000000
re-started bus-errors arbit-lost error-warn error-pass bus-off
0 1 0 0 0 0
I can add this in v4 version of the patch (Above log has this patch
applied).
Looks good.
quoted
And please check bus-off as well (short-circuiting CAN low and high).
I have not been able to check the bus-off condition by (short-circuiting
CAN low and high). The tec error count remains at 128 when I short the
CAN low and high pins and the status never goes BUSOFF.
You also need to send a message and the short-circuit should be at the
connector of the sending host. What tranceiver is used? Do you know?
Wolfgang.
From: Wolfgang Grandegger <hidden> Date: 2017-03-14 21:24:03
Am 14.03.2017 um 19:08 schrieb Wolfgang Grandegger:
Hello Akshay,
Am 14.03.2017 um 17:20 schrieb Akshay Bhat:
quoted
Hi Wolfgang,
On 03/14/2017 08:11 AM, Wolfgang Grandegger wrote:
quoted
... snip ...
quoted
quoted
A few other things to check:
Run "cangen" and monitor the message with "candump -e
any,0:0,#FFFFFFF".
Then 1) disconnect the cable or 2) short-circuit CAN low and high
at the
connector. You should see error messages. After reconnection or
removing
the short-circuit (and bus-off recovery) the state should go back to
"active".
With the above sequence, candump reports "ERRORFRAME" with
protocol-violation{{}{acknowledge-slot}}, bus-error. On re-connecting
the cable the can state goes back to ACTIVE and I see the messages that
were in the queue being sent.
Do you get the ACK error also with berr-reporting off? Would be nice if
you could show a candump log here.
Below is a log for disconnecting and re-connecting CAN cable scenario:
(Note this is on a 4.1.18 kernel with RT patch)
root@imx6qrom5420b1:~# ip link set can0 up type can bitrate 1000000
berr-reporting on
root@imx6qrom5420b1:~# candump -e any,0:0,#FFFFFFF &
Here "tx-error-passiv" is packed with a bus error. What I'm looking for
are state change messages similar to:
can0 20000204 [8] 00 08 00 00 00 00 60 00 ERRORFRAME
controller-problem{tx-error-warning}
state-change{tx-error-warning}
error-counter-tx-rx{{96}{0}}
can0 20000204 [8] 00 30 00 00 00 00 80 00 ERRORFRAME
controller-problem{tx-error-passive}
state-change{tx-error-passive}
error-counter-tx-rx{{128}{0}
They should always come, even with "berr-reporting off".
quoted
write: No buffer space available
root@imx6qrom5420b1:~# ip -s -d link show can0
4: can0: <NOARP,UP,LOWER_UP,ECHO> mtu 16 qdisc pfifo_fast state UNKNOWN
mode DEFAULT group default qlen 10
link/can promiscuity 0
can <BERR-REPORTING> state ERROR-PASSIVE (berr-counter tx 128 rx 0)
restart-ms 0
bitrate 1000000 sample-point 0.750
tq 62 prop-seg 5 phase-seg1 6 phase-seg2 4 sjw 1
hi3110: tseg1 2..16 tseg2 2..8 sjw 1..4 brp 1..64 brp-inc 1
clock 16000000
re-started bus-errors arbit-lost error-warn error-pass bus-off
0 6 0 1 1 0
The error warning and passive counter increased , though. Also the bus
error should come in at a rather hight rate. Looking to the code, maybe
you need to test STATF to check for state changes (and not ERR).
Likely the ERR bits are only valid if the BUSERR bit in INTF is set.
After some more messages there should be also:
can0 20000200 [8] 00 40 00 00 00 00 5F 00 ERRORFRAME
state-change{back-to-error-active}
error-counter-tx-rx{{95}{0}}
For each message sent, the error counter decreases by 8.
quoted
root@imx6qrom5420b1:~# ip -s -d link show can0
4: can0: <NOARP,UP,LOWER_UP,ECHO> mtu 16 qdisc pfifo_fast state UNKNOWN
mode DEFAULT group default qlen 10
link/can promiscuity 0
can state ERROR-ACTIVE (berr-counter tx 117 rx 0) restart-ms 0
bitrate 1000000 sample-point 0.750
tq 62 prop-seg 5 phase-seg1 6 phase-seg2 4 sjw 1
hi3110: tseg1 2..16 tseg2 2..8 sjw 1..4 brp 1..64 brp-inc 1
clock 16000000
re-started bus-errors arbit-lost error-warn error-pass bus-off
0 1 0 0 0 0
I can add this in v4 version of the patch (Above log has this patch
applied).
Looks good.
quoted
quoted
And please check bus-off as well (short-circuiting CAN low and high).
I have not been able to check the bus-off condition by (short-circuiting
CAN low and high). The tec error count remains at 128 when I short the
CAN low and high pins and the status never goes BUSOFF.
You also need to send a message and the short-circuit should be at the
connector of the sending host. What tranceiver is used? Do you know?
You could try to set a different bit-rate on the other CAN controller.
Then try to send or receive messages.
Wolfgang.
Here "tx-error-passiv" is packed with a bus error. What I'm looking for are
state change messages similar to:
can0 20000204 [8] 00 08 00 00 00 00 60 00 ERRORFRAME
controller-problem{tx-error-warning}
state-change{tx-error-warning}
error-counter-tx-rx{{96}{0}}
can0 20000204 [8] 00 30 00 00 00 00 80 00 ERRORFRAME
controller-problem{tx-error-passive}
state-change{tx-error-passive}
error-counter-tx-rx{{128}{0}
They should always come, even with "berr-reporting off".
HI-3110 has only 1 bus error interrupt. There is no dedicated state
change interrupts like other controllers.
So here is my plan:
- Have the bus error interrupt always enabled
- If berr-reporting off, then have the isr checks/reports state changes
- if berr-reporting on, then have the isr checks/reports bus errors
and state changes (Does it make sense packing the error message, if
the ISR finds both bus and state changes?)
quoted
write: No buffer space available
root@imx6qrom5420b1:~# ip -s -d link show can0
4: can0: <NOARP,UP,LOWER_UP,ECHO> mtu 16 qdisc pfifo_fast state UNKNOWN
mode DEFAULT group default qlen 10
link/can promiscuity 0
can <BERR-REPORTING> state ERROR-PASSIVE (berr-counter tx 128 rx 0)
restart-ms 0
bitrate 1000000 sample-point 0.750
tq 62 prop-seg 5 phase-seg1 6 phase-seg2 4 sjw 1
hi3110: tseg1 2..16 tseg2 2..8 sjw 1..4 brp 1..64 brp-inc 1
clock 16000000
re-started bus-errors arbit-lost error-warn error-pass bus-off
0 6 0 1 1 0
The error warning and passive counter increased , though. Also the bus error
should come in at a rather hight rate. Looking to the code, maybe
you need to test STATF to check for state changes (and not ERR).
Apologize, just realized In the above case some error packets were
lost, because I forgot to set the CPU frequency to max. Will resend
the log.
..snip...
After some more messages there should be also:
can0 20000200 [8] 00 40 00 00 00 00 5F 00 ERRORFRAME
state-change{back-to-error-active}
error-counter-tx-rx{{95}{0}}
For each message sent, the error counter decreases by 8.
The HI-3110 controller decrements the error counter by 1 for every message sent.
The error count increments by 8 when there is an error.
quoted
root@imx6qrom5420b1:~# ip -s -d link show can0
4: can0: <NOARP,UP,LOWER_UP,ECHO> mtu 16 qdisc pfifo_fast state UNKNOWN
mode DEFAULT group default qlen 10
link/can promiscuity 0
can state ERROR-ACTIVE (berr-counter tx 117 rx 0) restart-ms 0
bitrate 1000000 sample-point 0.750
tq 62 prop-seg 5 phase-seg1 6 phase-seg2 4 sjw 1
hi3110: tseg1 2..16 tseg2 2..8 sjw 1..4 brp 1..64 brp-inc 1
clock 16000000
re-started bus-errors arbit-lost error-warn error-pass bus-off
0 1 0 0 0 0
Strange, some counters got lost.
This was a bug introduced when adding berr-reporting, have fixed in v4 patch.
quoted
I have not been able to check the bus-off condition by (short-circuiting
CAN low and high). The tec error count remains at 128 when I short the
CAN low and high pins and the status never goes BUSOFF.
You also need to send a message and the short-circuit should be at the
connector of the sending host. What tranceiver is used? Do you know?
ADM3054 transceiver is used with HI-3111. I connected the
HI-3111/ADM3054 board to kvaser leaf and ran "cangen -i can0" and
"candump -e any,0:0,#FFFFFFF" on the board. Removed the cable and
shorted the CAN_H/L pins coming out of ADM3054. I will try your
suggestion of using a different bit-rate on the Kvaser leaf instead.
I appreciate your continued feedback, it has helped significantly
improve the error handling of the driver. Looking back I should have
based it on sja1000 or flexcan driver.
Thanks,
Akshay
--
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
Here "tx-error-passiv" is packed with a bus error. What I'm looking for are
state change messages similar to:
can0 20000204 [8] 00 08 00 00 00 00 60 00 ERRORFRAME
controller-problem{tx-error-warning}
state-change{tx-error-warning}
error-counter-tx-rx{{96}{0}}
can0 20000204 [8] 00 30 00 00 00 00 80 00 ERRORFRAME
controller-problem{tx-error-passive}
state-change{tx-error-passive}
error-counter-tx-rx{{128}{0}
They should always come, even with "berr-reporting off".
HI-3110 has only 1 bus error interrupt. There is no dedicated state
change interrupts like other controllers.
To double check: Could you please read INTF, ERR and STATF at the
beginning of the ISR and print it out (using dev_dbg and fiends). Then
run a test with no cable connected and bus error reporting off.
Wolfgang.
Here "tx-error-passiv" is packed with a bus error. What I'm looking for are
state change messages similar to:
can0 20000204 [8] 00 08 00 00 00 00 60 00 ERRORFRAME
controller-problem{tx-error-warning}
state-change{tx-error-warning}
error-counter-tx-rx{{96}{0}}
can0 20000204 [8] 00 30 00 00 00 00 80 00 ERRORFRAME
controller-problem{tx-error-passive}
state-change{tx-error-passive}
error-counter-tx-rx{{128}{0}
They should always come, even with "berr-reporting off".
HI-3110 has only 1 bus error interrupt. There is no dedicated state
change interrupts like other controllers.
I have little hope! Some revision of the flexcan controller have the
same problem
So here is my plan:
- Have the bus error interrupt always enabled
- If berr-reporting off, then have the isr checks/reports state changes
Error state change messages should always be there. These are the
important one.
- if berr-reporting on, then have the isr checks/reports bus errors
and state changes (Does it make sense packing the error message, if
the ISR finds both bus and state changes?)
write: No buffer space available
root@imx6qrom5420b1:~# ip -s -d link show can0
4: can0: <NOARP,UP,LOWER_UP,ECHO> mtu 16 qdisc pfifo_fast state UNKNOWN
mode DEFAULT group default qlen 10
link/can promiscuity 0
can <BERR-REPORTING> state ERROR-PASSIVE (berr-counter tx 128 rx 0)
restart-ms 0
bitrate 1000000 sample-point 0.750
tq 62 prop-seg 5 phase-seg1 6 phase-seg2 4 sjw 1
hi3110: tseg1 2..16 tseg2 2..8 sjw 1..4 brp 1..64 brp-inc 1
clock 16000000
re-started bus-errors arbit-lost error-warn error-pass bus-off
0 6 0 1 1 0
The error warning and passive counter increased , though. Also the bus error
should come in at a rather hight rate. Looking to the code, maybe
you need to test STATF to check for state changes (and not ERR).
Apologize, just realized In the above case some error packets were
lost, because I forgot to set the CPU frequency to max. Will resend
the log.
..snip...
quoted
After some more messages there should be also:
can0 20000200 [8] 00 40 00 00 00 00 5F 00 ERRORFRAME
state-change{back-to-error-active}
error-counter-tx-rx{{95}{0}}
For each message sent, the error counter decreases by 8.
The HI-3110 controller decrements the error counter by 1 for every message sent.
The error count increments by 8 when there is an error.
root@imx6qrom5420b1:~# ip -s -d link show can0
4: can0: <NOARP,UP,LOWER_UP,ECHO> mtu 16 qdisc pfifo_fast state UNKNOWN
mode DEFAULT group default qlen 10
link/can promiscuity 0
can state ERROR-ACTIVE (berr-counter tx 117 rx 0) restart-ms 0
bitrate 1000000 sample-point 0.750
tq 62 prop-seg 5 phase-seg1 6 phase-seg2 4 sjw 1
hi3110: tseg1 2..16 tseg2 2..8 sjw 1..4 brp 1..64 brp-inc 1
clock 16000000
re-started bus-errors arbit-lost error-warn error-pass bus-off
0 1 0 0 0 0
Strange, some counters got lost.
This was a bug introduced when adding berr-reporting, have fixed in v4 patch.
quoted
quoted
I have not been able to check the bus-off condition by (short-circuiting
CAN low and high). The tec error count remains at 128 when I short the
CAN low and high pins and the status never goes BUSOFF.
You also need to send a message and the short-circuit should be at the
connector of the sending host. What tranceiver is used? Do you know?
ADM3054 transceiver is used with HI-3111. I connected the
HI-3111/ADM3054 board to kvaser leaf and ran "cangen -i can0" and
"candump -e any,0:0,#FFFFFFF" on the board. Removed the cable and
shorted the CAN_H/L pins coming out of ADM3054. I will try your
suggestion of using a different bit-rate on the Kvaser leaf instead.
I appreciate your continued feedback, it has helped significantly
improve the error handling of the driver. Looking back I should have
based it on sja1000 or flexcan driver.
Well, the SJA1000 is the reference concerning error reporting. It's very
detailed. Most of the error cases from
http://lxr.free-electrons.com/source/include/uapi/linux/can/error.h
are SJA1000 related. Most other CAN controllers report much less. And
the Flexcan driver handles a lot of different chip revisions and uses
mail boxes. It's not a good base for the Hi-311x.
Wolfgang.
Wolfgang.
Hi Wolfgang,
On 03/15/2017 05:42 AM, Wolfgang Grandegger wrote:
Hello Akshay,
..snip..
quoted
So here is my plan:
- Have the bus error interrupt always enabled
- If berr-reporting off, then have the isr checks/reports state changes
Error state change messages should always be there. These are the
important one.
quoted
- if berr-reporting on, then have the isr checks/reports bus errors
and state changes (Does it make sense packing the error message, if
the ISR finds both bus and state changes?)
I have fixed the driver to handle the error reporting. Also thanks for
your tip for generating bus-off by setting the host device at a
different CAN bit rate! Below are logs with the updated driver. Let me
know if you have any concerns, if not I will submit the v4 patch.
berr-reporting on case:
http://pastebin.com/qDRLERmW
berr-reporting off case:
http://pastebin.com/fUn3j7qU
Thanks,
Akshay
--
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
From: Wolfgang Grandegger <hidden> Date: 2017-03-16 20:02:14
Hello Akshay,
Am 16.03.2017 um 18:06 schrieb Akshay Bhat:
Hi Wolfgang,
On 03/15/2017 05:42 AM, Wolfgang Grandegger wrote:
quoted
Hello Akshay,
..snip..
quoted
quoted
So here is my plan:
- Have the bus error interrupt always enabled
- If berr-reporting off, then have the isr checks/reports state changes
Error state change messages should always be there. These are the
important one.
quoted
- if berr-reporting on, then have the isr checks/reports bus errors
and state changes (Does it make sense packing the error message, if
the ISR finds both bus and state changes?)
I have fixed the driver to handle the error reporting. Also thanks for
your tip for generating bus-off by setting the host device at a
different CAN bit rate! Below are logs with the updated driver. Let me
know if you have any concerns, if not I will submit the v4 patch.
berr-reporting on case:
http://pastebin.com/qDRLERmW
Looks much better now! There are message for state changes to error
warning and passive. Just the following message is not correct:
(000.200824) can0 20000004 [8] 00 40 00 00 00 00 5F 19 ERRORFRAME
controller-problem{}
error-counter-tx-rx{{95}{25}}
Sorry, forgot to mention... the function can_change_state() [1]
should be used for that purpose, if possible. It fixes the issue
above as well.
Ditto.
I just had another look to the manual and there is this undocumented
STATFE register at offset 0x1E. It's mentioned in some other parts of
the doc as interrupt enable register for STATF events. I would assume
the same bit layout than STATF. If you set bit 2 (BUSOFF), 3 (ERRP)
and 4 (ERRW), you may get interrupts. It's worth a try, I think. If
it works, it's the much better solution.
Wolfgang.
[1] http://lxr.free-electrons.com/ident?i=can_change_state
Wolfgang.
Hi Wolfgang,
On 03/16/2017 04:02 PM, Wolfgang Grandegger wrote:
Looks much better now! There are message for state changes to error
warning and passive. Just the following message is not correct:
(000.200824) can0 20000004 [8] 00 40 00 00 00 00 5F 19 ERRORFRAME
controller-problem{}
error-counter-tx-rx{{95}{25}}
Sorry, forgot to mention... the function can_change_state() [1]
should be used for that purpose, if possible. It fixes the issue
above as well.
The updated driver (the one used to create the above log) is using
can_change_state() function. data[1] 40 corresponds to
CAN_ERR_CRTL_ACTIVE, so looks correct? Could it be that the can-utils I
am using is old and not reporting state change?
Tentative v4 driver for reference:
http://pastebin.com/4xFVL1Sj
Ditto.
I just had another look to the manual and there is this undocumented
STATFE register at offset 0x1E. It's mentioned in some other parts of
the doc as interrupt enable register for STATF events. I would assume
the same bit layout than STATF. If you set bit 2 (BUSOFF), 3 (ERRP)
and 4 (ERRW), you may get interrupts. It's worth a try, I think. If
it works, it's the much better solution.
The HI-311x has a INT pin and a STAT pin. The hardware I have has only
the INT pin connected to the processor. If the STAT pin was also
connected, then like you mentioned it could be a much better solution to
use STAT for state changes.
Enabling BUSOFF/ERRP/ERRW bits in STATFE did not generate any interrupts
on the INT pin. Should we make it a requirement that both INT and STAT
pins need to be connected in hardware for the driver to do the error
reporting?
Thanks,
Akshay
From: Wolfgang Grandegger <hidden> Date: 2017-03-17 07:39:36
Hello Akshay,
Am 16.03.2017 um 23:29 schrieb Akshay Bhat:
Hi Wolfgang,
On 03/16/2017 04:02 PM, Wolfgang Grandegger wrote:
quoted
Looks much better now! There are message for state changes to error
warning and passive. Just the following message is not correct:
(000.200824) can0 20000004 [8] 00 40 00 00 00 00 5F 19 ERRORFRAME
controller-problem{}
error-counter-tx-rx{{95}{25}}
Sorry, forgot to mention... the function can_change_state() [1]
should be used for that purpose, if possible. It fixes the issue
above as well.
The updated driver (the one used to create the above log) is using
can_change_state() function. data[1] 40 corresponds to
CAN_ERR_CRTL_ACTIVE, so looks correct? Could it be that the can-utils I
am using is old and not reporting state change?
Hm, yes. The raw data looks correct. You could download and build a
recent version from "https://github.com/linux-can/can-utils" to check.
It could also be a bug.
Ditto.
I just had another look to the manual and there is this undocumented
STATFE register at offset 0x1E. It's mentioned in some other parts of
the doc as interrupt enable register for STATF events. I would assume
the same bit layout than STATF. If you set bit 2 (BUSOFF), 3 (ERRP)
and 4 (ERRW), you may get interrupts. It's worth a try, I think. If
it works, it's the much better solution.
The HI-311x has a INT pin and a STAT pin. The hardware I have has only
the INT pin connected to the processor. If the STAT pin was also
connected, then like you mentioned it could be a much better solution to
use STAT for state changes.
OK, I understand.
Enabling BUSOFF/ERRP/ERRW bits in STATFE did not generate any interrupts
on the INT pin. Should we make it a requirement that both INT and STAT
pins need to be connected in hardware for the driver to do the error
reporting?
As I said, it's the better solution, especially if interrupt flooding
does harm. How does your system behave when bus errors come in due to no
cable connected?
So far using NAPI was mandatory. There is the problem of out-of-order
message reception if handled in the isr on multi processor systems.
Marc, what is the current policy?
Wolfgang.
From: Wolfgang Grandegger <hidden> Date: 2017-03-17 08:18:56
Hello Akshay,
Am 17.03.2017 um 08:39 schrieb Wolfgang Grandegger:
Hello Akshay,
Am 16.03.2017 um 23:29 schrieb Akshay Bhat:
quoted
Hi Wolfgang,
On 03/16/2017 04:02 PM, Wolfgang Grandegger wrote:
quoted
Looks much better now! There are message for state changes to error
warning and passive. Just the following message is not correct:
(000.200824) can0 20000004 [8] 00 40 00 00 00 00 5F 19
ERRORFRAME
controller-problem{}
error-counter-tx-rx{{95}{25}}
Sorry, forgot to mention... the function can_change_state() [1]
should be used for that purpose, if possible. It fixes the issue
above as well.
The updated driver (the one used to create the above log) is using
can_change_state() function. data[1] 40 corresponds to
CAN_ERR_CRTL_ACTIVE, so looks correct? Could it be that the can-utils I
am using is old and not reporting state change?
Hm, yes. The raw data looks correct. You could download and build a
recent version from "https://github.com/linux-can/can-utils" to check.
It could also be a bug.
Support for that flags has been added in December 2014.
Wolfgang-
Hi Wolfgang,
On 03/17/2017 03:39 AM, Wolfgang Grandegger wrote:
Hello Akshay,
Am 16.03.2017 um 23:29 schrieb Akshay Bhat:
quoted
Hi Wolfgang,
On 03/16/2017 04:02 PM, Wolfgang Grandegger wrote:
quoted
Looks much better now! There are message for state changes to error
warning and passive. Just the following message is not correct:
(000.200824) can0 20000004 [8] 00 40 00 00 00 00 5F 19
ERRORFRAME
controller-problem{}
error-counter-tx-rx{{95}{25}}
Sorry, forgot to mention... the function can_change_state() [1]
should be used for that purpose, if possible. It fixes the issue
above as well.
The updated driver (the one used to create the above log) is using
can_change_state() function. data[1] 40 corresponds to
CAN_ERR_CRTL_ACTIVE, so looks correct? Could it be that the can-utils I
am using is old and not reporting state change?
Hm, yes. The raw data looks correct. You could download and build a
recent version from "https://github.com/linux-can/can-utils" to check.
It could also be a bug.
Turned out to be a old version of can-utils. Using the above git tree
reports the flag.
(000.200308) can0 20000004 [8] 00 40 00 00 00 00 5F 00 ERRORFRAME
controller-problem{back-to-error-active}
error-counter-tx-rx{{95}{0}}
quoted
Enabling BUSOFF/ERRP/ERRW bits in STATFE did not generate any interrupts
on the INT pin. Should we make it a requirement that both INT and STAT
pins need to be connected in hardware for the driver to do the error
reporting?
As I said, it's the better solution, especially if interrupt flooding
does harm. How does your system behave when bus errors come in due to no
cable connected?
I did not see any issues on the system with the cable disconnected. In
my particular setup with the cable disconnected the system goes to
tx-error-passive and does not get any further interrupts until a state
change occurs.
So far using NAPI was mandatory. There is the problem of out-of-order
message reception if handled in the isr on multi processor systems.
Marc, what is the current policy?
Since this is a SPI based CAN, I am wary for any additional latencies
NAPI might introduce. The RX handling is being done at the very
beginning of the ISR for this reason.
Can we go ahead with the existing implementation and re-visit this at a
later time?
Thanks again for all your help in reviewing/improving the driver :)
Akshay
From: Wolfgang Grandegger <hidden> Date: 2017-03-17 17:35:11
Hi Akshay,
Am 17.03.2017 um 17:00 schrieb Akshay Bhat:
Hi Wolfgang,
On 03/17/2017 03:39 AM, Wolfgang Grandegger wrote:
quoted
Hello Akshay,
Am 16.03.2017 um 23:29 schrieb Akshay Bhat:
quoted
Hi Wolfgang,
On 03/16/2017 04:02 PM, Wolfgang Grandegger wrote:
quoted
Looks much better now! There are message for state changes to error
warning and passive. Just the following message is not correct:
(000.200824) can0 20000004 [8] 00 40 00 00 00 00 5F 19
ERRORFRAME
controller-problem{}
error-counter-tx-rx{{95}{25}}
Sorry, forgot to mention... the function can_change_state() [1]
should be used for that purpose, if possible. It fixes the issue
above as well.
The updated driver (the one used to create the above log) is using
can_change_state() function. data[1] 40 corresponds to
CAN_ERR_CRTL_ACTIVE, so looks correct? Could it be that the can-utils I
am using is old and not reporting state change?
Hm, yes. The raw data looks correct. You could download and build a
recent version from "https://github.com/linux-can/can-utils" to check.
It could also be a bug.
Turned out to be a old version of can-utils. Using the above git tree
reports the flag.
(000.200308) can0 20000004 [8] 00 40 00 00 00 00 5F 00 ERRORFRAME
controller-problem{back-to-error-active}
error-counter-tx-rx{{95}{0}}
quoted
quoted
Enabling BUSOFF/ERRP/ERRW bits in STATFE did not generate any interrupts
on the INT pin. Should we make it a requirement that both INT and STAT
pins need to be connected in hardware for the driver to do the error
reporting?
As I said, it's the better solution, especially if interrupt flooding
does harm. How does your system behave when bus errors come in due to no
cable connected?
I did not see any issues on the system with the cable disconnected. In
my particular setup with the cable disconnected the system goes to
tx-error-passive and does not get any further interrupts until a state
change occurs.
Hm, that's unusual. Cable disconnected and then send a message:
$ grep /proc/interrupts; sleep 10; /proc/interrupts
should make things clear. But maybe it's a clever chip and it does stop
sending error messages if the error counter does not change any more.
After bus-off, the chip is quiet, of course. Should have a closer look
to the CAN standard.
quoted
So far using NAPI was mandatory. There is the problem of out-of-order
message reception if handled in the isr on multi processor systems.
Marc, what is the current policy?
Since this is a SPI based CAN, I am wary for any additional latencies
NAPI might introduce. The RX handling is being done at the very
beginning of the ISR for this reason.
Can we go ahead with the existing implementation and re-visit this at a
later time?
Likely yes, as Marc has already reviewed the driver once.
BTW: what system board/processor are you using?
Thanks again for all your help in reviewing/improving the driver :)
Hi Wolfgang,
On 03/17/2017 01:04 PM, Wolfgang Grandegger wrote:
Hm, that's unusual. Cable disconnected and then send a message:
$ grep /proc/interrupts; sleep 10; /proc/interrupts
should make things clear. But maybe it's a clever chip and it does stop
sending error messages if the error counter does not change any more.
After bus-off, the chip is quiet, of course. Should have a closer look
to the CAN standard.
So far using NAPI was mandatory. There is the problem of out-of-order
message reception if handled in the isr on multi processor systems.
Marc, what is the current policy?
Since this is a SPI based CAN, I am wary for any additional latencies
NAPI might introduce. The RX handling is being done at the very
beginning of the ISR for this reason.
Can we go ahead with the existing implementation and re-visit this at a
later time?
Likely yes, as Marc has already reviewed the driver once.
Thanks, I will go ahead and submit v4 patch.
BTW: what system board/processor are you using?
It is a custom board using Phytec phyFLEX-i.MX6 Dual SOM.
--
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
From: Wolfgang Grandegger <hidden> Date: 2017-03-18 12:31:20
Hello Akshay,
Am 17.03.2017 um 19:28 schrieb Akshay Bhat:
Hi Wolfgang,
On 03/17/2017 01:04 PM, Wolfgang Grandegger wrote:
quoted
Hm, that's unusual. Cable disconnected and then send a message:
$ grep /proc/interrupts; sleep 10; /proc/interrupts
should make things clear. But maybe it's a clever chip and it does stop
sending error messages if the error counter does not change any more.
After bus-off, the chip is quiet, of course. Should have a closer look
to the CAN standard.
OK, then there is no good reason connecting the STAT interrupt pin.
Wolfgang.
--
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