D_CAN interrupts problem

Jiewen Wang

Other Parts Discussed in Thread: DM385

We have a project is going to use CAN bus to communicate with some handheld device. We currently working on DM814x-EVM with ezsdk 5.04.00.11 in Linux environment. I have libsocketcan and canutils as test program. I have seen the d-can drive and can protocol were up during boot time.

CAN device driver interface

CAN bus driver for Bosch D_CAN controller 1.0

d_can d_can.0: pdev c052bec0 pdev->dev c052bec8 offset 0 clk dcan0_fck

d_can d_can.0: mem 0x481cc000

d_can d_can.0: d_can device registered (ndev= cc98b000, irq=52, irq_obj=53) base=0xfa1cc000

The steps that I run my test on the target were:

ip link set can0 type can bitrate 50000
ip link set can0 up
candump can0 &
cansend can0 -i 0x01 0xaa 0xbb

My expectation is I send a data frame on can0 that I will receive it on can0 with candump. I used the d_can driver from the DevKit, I get status interrupts, but no message object interrupts. Then I set the D_CAN_IF_MCTL_INTPND bit in D_CAN_IFMCTL register, this way I get tx interrupts. When I probe the CAN_H and CAN_L, I did see the data. However, I never get any rx interrupts.

My question are:

Is this the right way to test can bus?
Base on the DM814x TRM I've read ID_CAN_IF_MCTL_INTPND bit will be set when data frame is transferred, why do I need to set it in order to get tx interrupts?
How do I get rx interrupts to work?

If anybody could provide any comments or suggestions, I would appreciate very much.

Jiewen

over 11 years ago

0 Mark Chambers over 11 years ago

Prodigy 70 points

I am also trying to get CAN to work on a DM814x. I also can transmit data but not receive anything. I put my oscilloscope on the DCAN0_RX pin (AG6) and it appears to be stuck high, like it's driving the line instead of receiving. Have you looked at this?

0 Jiewen Wang over 11 years ago in reply to Mark Chambers

Prodigy 250 points

Hi Mark,

I probed the DCAN0_RX pin, it looks matching the CAN_H and CAN_L signals. it is not always high. Have you made any changes in the driver?

Jiewen

0 Mark Chambers over 11 years ago in reply to Jiewen Wang

Prodigy 70 points

Thank you for that information. I think I may have a hardware problem. My CAN PHY chip, SN64HVD251 is a 5V part, I wonder if the Rx damaged the 8148 pin?

I did move the pinmuxing to a later point because my hardware was driving the CAN bus on boot, but other than that I did not change anything in software.

0 Jiewen Wang over 11 years ago in reply to Mark Chambers

Prodigy 250 points

I did commented out the following pinmux setting in mux814x.c

_TI814X_MUXENTRY(UART0_RTSN, 0, "uart0_rtsn", "sata_act1_led_mux", NULL, "dcan1_rx", "spi1_cs2", NULL, "mmc2_sdcd", NULL), /* DM385 only */

0 Mark Chambers over 11 years ago in reply to Jiewen Wang

Prodigy 70 points

It turned out I had a bad solder connection! So I am now able to use candump to see data on my car's OBD connector. No changes to the driver, but this is custom hardware, not the EVM.

0 Anil over 11 years ago in reply to Jiewen Wang

TI__Expert 6655 points

Hi Jiewen,

Don't change any part of the code from release w.r.t to D_CAN and follow these steps?

1. Board-Board connection details

http://processors.wiki.ti.com/index.php/TI81XX_DCAN_Driver_Guide#dm814x

2. Can you use exact steps mentioned at

http://processors.wiki.ti.com/index.php/TI81XX_DCAN_Driver_Guide#Quick_Steps

3. D_CAN0 don't have relation with these changes

Jiewen Wang said:

I did commented out the following pinmux setting in mux814x.c

_TI814X_MUXENTRY(UART0_RTSN, 0, "uart0_rtsn", "sata_act1_led_mux", NULL, "dcan1_rx", "spi1_cs2", NULL, "mmc2_sdcd", NULL), /* DM385 only */

Regards

AnilKumar

Please mark this Forum post as answered via the Verify Answer button below if it helps answer your question. Thanks!

0 Jiewen Wang over 11 years ago in reply to Anil

Prodigy 250 points

AnilKumar,

I did a test as in the TI81XX_DCAN_Driver_Guide except I connected a known CAN device to the dcan0. The kernel is the original one. Test failed, still no tx or rx interrupt occurred. I did see data on CANH and CANL though. Here is the log:

root@dm814x-evm:~# ip link set can0 type can bitrate 50000 triple-sampling on loopback on
root@dm814x-evm:~# ip link set can0 up
d_can d_can: can0: setting CAN BT = 0x1c18
root@dm814x-evm:~# candump can0 &
root@dm814x-evm:~# open 0 'can0'.
using interface name 'can0'.

root@dm814x-evm:~# cansend can0 -i 0x4e1 1 1
interface = can0, family = 29, type = 3, proto = 1, idx = 2

root@dm814x-evm:~# cat /proc/interrupts
CPU0
4: 0 INTC omap2_elm
12: 236 INTC edma
14: 0 INTC edma_error
16: 0 INTC ahci
17: 0 INTC cppi41_dma
18: 1 INTC musb-hdrc.0
19: 1 INTC musb-hdrc.1
20: 0 INTC gpmc
28: 729 INTC mmc1
30: 1 INTC omap_i2c
37: 0 INTC SGX ISR
38: 0 INTC HDMI
40: 0 INTC cpsw.0
41: 67104 INTC cpsw.0
42: 69946 INTC cpsw.0
43: 0 INTC cpsw.0
44: 0 INTC serial idle
45: 0 INTC serial idle
46: 0 INTC serial idle
52: 2 INTC can0
53: 0 INTC can0
64: 13 INTC mmc0
67: 6953 INTC gp timer
70: 774 INTC omap_i2c
72: 604 INTC serial idle, OMAP UART0
73: 0 INTC serial idle
74: 0 INTC serial idle
75: 0 INTC rtc0
76: 0 INTC rtc0
77: 9789 INTC mailbox-dsp, mailbox-video, mailbox-vpss
122: 0 INTC omap-iommu.1
123: 0 INTC omap-iommu.0
191: 2 GPIO qt602240_ts
Err: 0
root@dm814x-evm:~# cat /proc/net/can/stats

1 transmitted frames (TXF)
0 received frames (RXF)
0 matched frames (RXMF)

0 % total match ratio (RXMR)
0 frames/s total tx rate (TXR)
0 frames/s total rx rate (RXR)

0 % current match ratio (CRXMR)
0 frames/s current tx rate (CTXR)
0 frames/s current rx rate (CRXR)

0 % max match ratio (MRXMR)
1 frames/s max tx rate (MTXR)
0 frames/s max rx rate (MRXR)

1 current receive list entries (CRCV)
2 maximum receive list entries (MRCV)

root@dm814x-evm:~# ip -s -d link show
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
RX: bytes packets errors dropped overrun mcast
340 6 0 0 0 0
TX: bytes packets errors dropped carrier collsns
340 6 0 0 0 0
2: can0: <NOARP,UP,LOWER_UP,ECHO> mtu 16 qdisc pfifo_fast state UNKNOWN qlen 10
link/can
can <LOOPBACK,TRIPLE-SAMPLING> state ERROR-ACTIVE (berr-counter tx 0 rx 0) restart-ms 0
bitrate 50000 sample-point 0.875
tq 1250 prop-seg 6 phase-seg1 7 phase-seg2 2 sjw 1
d_can: tseg1 1..16 tseg2 1..8 sjw 1..4 brp 1..1024 brp-inc 1
clock 20000000
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
0 0 0 0 0 0
3: eth0: <BROADCAST,MULTICAST,ALLMULTI,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 40:5f:c2:08:94:18 brd ff:ff:ff:ff:ff:ff
RX: bytes packets errors dropped overrun mcast
92006778 68061 0 93 0 0
TX: bytes packets errors dropped carrier collsns
3657904 39452 0 0 0 0
root@dm814x-evm:~#

Regards.

Jiewen

0 Anil over 11 years ago in reply to Jiewen Wang

TI__Expert 6655 points

Hi Jiewen,

Is this issue resolved?

Thanks

AnilKumar

0 Ershad Hussein over 11 years ago in reply to Anil

Prodigy 55 points

Hi Anil,

I do see the same problem as Jiewen. Is it a connector solder issue? Or a current loop problem requiring modifying the diff term resistor? Or a S/W issue?

Best regards,

Ershad

0 Anil over 11 years ago in reply to Ershad Hussein

TI__Expert 6655 points

Hi Ershad,

Can you get these details to get more idea on what is happening really?

1. cat /proc/interrupts | grep -i can

2. ip -d -s link show can0

3. cat /proc/net/can/stats

4. Checking the error details in dcan registers

a. devmem2 0x481CC000

b. devmem2 0x481CC004

Note: devmem2 is a utility used to read data

Regards

AnilKumar

Please mark this Forum post as answered via the Verify Answer button below if it helps answer your question. Thanks!

0 Ershad Hussein over 11 years ago in reply to Anil

Prodigy 55 points

Hi Anil, here is what you asked for

root@dm814x-evm:~# ./CANBringup.sh

can0 state: STOPPED

can0 bitrate: 1000000, sample-point: 0.750

d_can d_can: can0: setting CAN BT = 0x4d00

root@dm814x-evm:~# cat /proc/interrupts | grep -i can

52: 2 INTC can0

53: 0 INTC can0

root@dm814x-evm:~# ip -d -s link show can0

2: can0: <NOARP,UP,LOWER_UP,ECHO> mtu 16 qdisc pfifo_fast state UNKNOWN qlen 10

link/can

can state ERROR-ACTIVE (berr-counter tx 0 rx 0) restart-ms 0

bitrate 1000000 sample-point 0.750

tq 50 prop-seg 7 phase-seg1 7 phase-seg2 5 sjw 1

d_can: tseg1 1..16 tseg2 1..8 sjw 1..4 brp 1..1024 brp-inc 1

clock 20000000

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

0 0 0 0 0 0

root@dm814x-evm:~# cat /proc/net/can/stats

0 transmitted frames (TXF)

0 received frames (RXF)

0 matched frames (RXMF)

0 % total match ratio (RXMR)

0 frames/s total tx rate (TXR)

0 frames/s total rx rate (RXR)

0 % current match ratio (CRXMR)

0 frames/s current tx rate (CTXR)

0 frames/s current rx rate (CRXR)

0 % max match ratio (MRXMR)

0 frames/s max tx rate (MTXR)

0 frames/s max rx rate (MRXR)

0 current receive list entries (CRCV)

0 maximum receive list entries (MRCV)

root@dm814x-evm:~# devmem2 0x481CC000

/dev/mem opened.

Memory mapped at address 0x40209000.

Read at address 0x481CC000 (0x40209000): 0x00000000

root@dm814x-evm:~# devmem2 0x481CC004

/dev/mem opened.

Memory mapped at address 0x4015d000.

Read at address 0x481CC004 (0x4015d004): 0x00000107

root@dm814x-evm:~#

Best regards,

Ershad

0 Ershad Hussein over 11 years ago in reply to Ershad Hussein

Prodigy 55 points

Hi Anil,

I got the candump to working. However, there is still an intermittent issue, suggesting perhaps a hardware (pin contact or impedance) problem.

Best regards,

Ershad

0 Ershad Hussein over 11 years ago in reply to Ershad Hussein

Prodigy 55 points

Hi again Anil,

I can not rule out a S/W problem, because if I execute the following steps, I can get the candump working reliably:

canconfig can0 stop
canconfig can0 bitrate 1000000
ip link set can0 up
cansend can0 -i 0x001 0x12 0x23 0x34 0x56 0x45 0x78 0xAB 0xFF
canconfig can0 stop
canconfig can0 bitrate 125000
ip link set can0 up

0 Anil over 11 years ago in reply to Ershad Hussein

TI__Expert 6655 points

Hi Ershad Hussein,

root@dm814x-evm:~# devmem2 0x481CC000

/dev/mem opened.

Memory mapped at address 0x40209000.

Read at address 0x481CC000 (0x40209000): 0x00000000 (This is wrong this should be 0x0000000e)

Can you change the D_CAN driver similar to this patch and see whether the issue got fixed or not (I hope this is the fix)? Commit message has the details of the issue.

http://git.kernel.org/?p=linux/kernel/git/next/linux-next.git;a=commitdiff;h=f461f27a4436dbe691908fe08b867ef888848cc3;hp=148c87c89e1a8863d3d965179f3ab1a06490569e

Thanks

AnilKumar

Please mark this Forum post as answered via the Verify Answer button below if it helps answer your question. Thanks!

0 Ershad Hussein over 11 years ago in reply to Anil

Prodigy 55 points

Thanks Anil, it fixes the problem.

0 Jiewen Wang over 11 years ago in reply to Ershad Hussein

Prodigy 250 points

Antil,

I have a new problem now. Sometimes I see DCAN bus stuck in warning/error passive state, 0x481cc004 = 0x00000063, so once it gets in this state, it is unable to recover. I have to restart the system. I'd like to know if there is a way it can be recovered by software. (I have incorporated the change you mentioned above)

root@dm814x-evm:~/dm814x# devmem2 0x481cc000
/dev/mem opened.
Memory mapped at address 0x40297000.
Read at address 0x481CC000 (0x40297000): 0x0002000E

root@dm814x-evm:~/dm814x# devmem2 0x481cc004
/dev/mem opened.
Memory mapped at address 0x4003b000.
Read at address 0x481CC004 (0x4003b004): 0x00000067

root@dm814x-evm:~/dm814x# devmem2 0x481cc008
/dev/mem opened.
Memory mapped at address 0x4033a000.
Read at address 0x481CC008 (0x4033a008): 0x00000080

Thanks.

Jiewen

Processors

Processors forum

D_CAN interrupts problem