Skip to content

mcp2515 on spi1 can0 not working #5346

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Closed
martburg opened this issue Feb 6, 2023 · 28 comments
Closed

mcp2515 on spi1 can0 not working #5346

martburg opened this issue Feb 6, 2023 · 28 comments

Comments

@martburg
Copy link

martburg commented Feb 6, 2023 •
edited by pelwell
Loading

Describe the bug

mcp2515 on spi1 - candump can0 not generating output

Steps to reproduce the behaviour

waveshare 2-ch-can-hat connected to raspberry B3+

INT1 --- Pin 29 GPIO 5
INT0 --- Pin 31 GPIO 6
CS1 ---- Pin 11 GPIO 17 (SPI1 CE1)
CS2 ---- Pin 12 GPIO 18 (SPI1 CE0)
SCK ---- Pin 40 GPIO21 (SPI1 SCLK)
MOSI -- Pin 38 GPIO20 (SPI1 MOSI)
MISO -- Pin 35 GPIO19 (SPI1 MISO)
GND --- Pin 6 GND
5V ----- Pin 1 5V

/boot/config.txt:

dtoverlay=spi1-2cs,cs0_spidev=off,cs1_spidev=off
dtoverlay=t-sp1-can0,oscillator=16000000,interrupt=5
dtoverlay=t-sp1-can1,oscillator=16000000,interrupt=6

t-sp1-can0.dtbo compiles from:

 /*  
*Device tree overlay for mcp251x/can0 on spi1.0
 */

/dts-v1/;
/plugin/;

/ {
    compatible = "brcm,bcm2835", "brcm,bcm2836", "brcm,bcm2708", "brcm,bcm2709";
    /* disable spi-dev for spi1.0 */
    fragment@0 {
        target = <&spi1>;
        __overlay__ {
            status = "okay";
        };
    };

    fragment@1 {
	target = <&spidev0>;
	__overlay__ {
	    status = "disabled";
	};
    };

    /* the interrupt pin of the can-controller */
    fragment@2 {
        target = <&gpio>;
        __overlay__ {
            can0_pins: can0_pins {
                brcm,pins = <25>;
                brcm,function = <0>; /* input */
            };
        };
    };

    /* the clock/oscillator of the can-controller */
    fragment@3 {
        target-path = "/clocks";
        __overlay__ {
            /* external oscillator of mcp2515 on SPI0.0 */
            can0_osc: can0_osc {
                compatible = "fixed-clock";
                #clock-cells = <0>;
                clock-frequency  = <16000000>;
            };
        };
    };

    /* the spi config of the can-controller itself binding everything together */
    fragment@4 {
        target = <&spi1>;
        __overlay__ {
            /* needed to avoid dtc warning */
            #address-cells = <1>;
            #size-cells = <0>;
            can0: mcp2515@0 {
                reg = <0>;
                compatible = "microchip,mcp2515";
                pinctrl-names = "default";
                pinctrl-0 = <&can0_pins>;
                spi-max-frequency = <10000000>;
                interrupt-parent = <&gpio>;
                interrupts = <25 0x2>;
                clocks = <&can0_osc>;
            };
        };
    };
    __overrides__ {
        oscillator = <&can0_osc>,"clock-frequency:0";
        spimaxfrequency = <&can0>,"spi-max-frequency:0";
        interrupt = <&can0_pins>,"brcm,pins:0",<&can0>,"interrupts:0";
    };
};

t-sp1-can1.dtbo compiles from:

/*
 * Device tree overlay for mcp251x/can1 on spi1.1 edited by petit_miner
 */

/dts-v1/;
/plugin/;

/ {
    compatible = "brcm,bcm2835", "brcm,bcm2836", "brcm,bcm2708", "brcm,bcm2709";
    /* disable spi-dev for spi1.1 */
    fragment@0 {
        target = <&spi1>;
        __overlay__ {
            status = "okay";
        };
    };

    fragment@1 {
	target = <&spidev1>;
	__overlay__ {
	    status = "disabled";
	};
    };

    /* the interrupt pin of the can-controller */
    fragment@2 {
        target = <&gpio>;
        __overlay__ {
            can1_pins: can1_pins {
                brcm,pins = <25>;
                brcm,function = <0>; /* input */
            };
        };
    };

    /* the clock/oscillator of the can-controller */
    fragment@3 {
        target-path = "/clocks";
        __overlay__ {
            /* external oscillator of mcp2515 on spi0.1 */
            can1_osc: can1_osc {
                compatible = "fixed-clock";
                #clock-cells = <0>;
                clock-frequency  = <16000000>;
            };
        };
    };

    /* the spi config of the can-controller itself binding everything together */
    fragment@4 {
        target = <&spi1>;
        __overlay__ {
            /* needed to avoid dtc warning */
            #address-cells = <1>;
            #size-cells = <0>;
            can1: mcp2515@1 {
                reg = <1>;
                compatible = "microchip,mcp2515";
                pinctrl-names = "default";
                pinctrl-0 = <&can1_pins>;
                spi-max-frequency = <10000000>;
                interrupt-parent = <&gpio>;
                interrupts = <25 0x2>;
                clocks = <&can1_osc>;
            };
        };
    };
    __overrides__ {
        oscillator = <&can1_osc>,"clock-frequency:0";
        spimaxfrequency = <&can1>,"spi-max-frequency:0";
        interrupt = <&can1_pins>,"brcm,pins:0",<&can1>,"interrupts:0";
    };
};

dmesg|grep spi

[ 8.232885] mcp251x spi1.1 can0: MCP2515 successfully initialized.
[ 8.245989] mcp251x spi1.0 can1: MCP2515 successfully initialized.

spi1.1 linked to can0 !
spi1.0 linked to can1 !

Device (s)

Raspberry Pi 1 Mod. B+

System

Raspberry Pi reference 2021-01-11
Generated using pi-gen, https://github.com/RPi-Distro/pi-gen, 21090519d85bdaa1615d5d5057d37b09368ea5d2, stage2

Oct 29 2021 10:49:08
Copyright (c) 2012 Broadcom
version b8a114e5a9877e91ca8f26d1a5ce904b2ad3cf13 (clean) (release) (start)

Linux raspberrypi 5.10.63-v7+ #1459 SMP Wed Oct 6 16:41:10 BST 2021 armv7l GNU/Linux

Logs

No response

Additional context

No response
@pelwell
Copy link
Contributor

pelwell commented Feb 6, 2023

spi1.1 linked to can0 !
spi1.0 linked to can1 !

Is that your bug report? Please explain clearly what is not working.

@martburg
Copy link
Author

martburg commented Feb 6, 2023

The Bug:

candump can0 is not producing any output.

and of course a can signal is applied to can0

I'm not sure about the linking of spi1.1 to can 0 and spi1.0 to can1 it's an observation maybe a clue to find the bug

@pelwell
Copy link
Contributor

pelwell commented Feb 6, 2023

  1. Does it work with just one of the overlays applied?
  2. Does /proc/interrupts show any numbers increasing (it might be labelled as something like gpio 5/6)?
  3. Does dmesg -l err,warn show anything relevant?

Without this board and some CAN hardware my support is going to be limited.

@martburg
Copy link
Author

martburg commented Feb 6, 2023

will check tomorow

thx

@pelwell
Copy link
Contributor

pelwell commented Feb 6, 2023

  1. Is the bus properly terminated at both ends? I would hope the WaveShare board does one end.

@martburg
Copy link
Author

martburg commented Feb 6, 2023

Yes, the bus is terminated

when I connect the same CAN lines to CAN1 --candump can1-- is OK

@martburg
Copy link
Author

martburg commented Feb 7, 2023 •
edited by pelwell
Loading

@pelwell
ad 1. dtoverlay=t-sp1-can0,oscillator=16000000,interrupt=5

           dmesg|grep spi ----> [    8.109524] mcp251x spi1.0: Cannot initialize MCP2515. Wrong wiring?
                                            [    8.109766] mcp251x spi1.0: Probe failed, err=19

         dtoverlay=t-sp1-can1,oscillator=16000000,interrupt=6   
           dmesg|grep spi ----> [    8.109524] mcp251x spi1.1: Cannot initialize MCP2515. Wrong wiring?
                                            [    8.109766] mcp251x spi1.1: Probe failed, err=19

with both overlays can1 is OK and can0 not OK

ad 2. dtoverlay=t-sp1-can0,oscillator=16000000,interrupt=5

         cat /proc/interrupts ---->  
          CPU0       CPU1       CPU2       CPU3       
 50:        1592             0            0          0  ARMCTRL-level   1 Edge      3f00b880.mailbox
 51:          144             0            0          0  ARMCTRL-level   2 Edge      VCHIQ doorbell
 73:              0             0            0          0  ARMCTRL-level  48 Edge      bcm2708_fb DMA
 75:          213             0            0          0  ARMCTRL-level  50 Edge      DMA IRQ
 77:      10376             0            0          0  ARMCTRL-level  52 Edge      DMA IRQ
 78:              0             0            0          0  ARMCTRL-level  53 Edge      DMA IRQ
 81:              0             0            0          0  ARMCTRL-level  56 Edge      DMA IRQ
 86:              0             0            0          0  ARMCTRL-level  61 Edge      3f215080.spi
 89:      95250             0            0          0  ARMCTRL-level  64 Edge      dwc_otg, dwc_otg_pcd, dwc_otg_hcd:usb1
111:            0              0            0          0  ARMCTRL-level  86 Edge      3f204000.spi
113:        531              0            0          0  ARMCTRL-level  88 Edge      mmc0
114:       4219             0            0          0  ARMCTRL-level  89 Edge      uart-pl011
119:       3536             0            0          0  ARMCTRL-level  94 Edge      mmc1
194:            0              0             0          0  bcm2836-timer   0 Edge      arch_timer
195:       7536     10754      18405      11109  bcm2836-timer   1 Edge      arch_timer
198:            0              0              0          0  bcm2836-pmu   9 Edge      arm-pmu
FIQ:              usb_fiq
IPI0:            0              0             0             0  CPU wakeup interrupts
IPI1:            0              0             0              0  Timer broadcast interrupts
IPI2:         470         484          402          463  Rescheduling interrupts
IPI3:       8269      14516      11565       8577  Function call interrupts
IPI4:            0               0             0              0  CPU stop interrupts
IPI5:        248           453       1030          644  IRQ work interrupts
IPI6:            0               0             0              0  completion interrupts
Err:             0 

dmesg-l err,warn ---->
[    0.000000] Kernel parameter elevator= does not have any effect anymore.
               Please use sysfs to set IO scheduler for individual devices.
[    2.735317] Core Release: 2.80a
[    2.735334] Setting default values for core params
[    2.735366] Finished setting default values for core params
[    2.935758] Using Buffer DMA mode
[    2.935772] Periodic Transfer Interrupt Enhancement - disabled
[    2.935784] Multiprocessor Interrupt Enhancement - disabled
[    2.935797] OTG VER PARAM: 0, OTG VER FLAG: 0
[    2.935825] Dedicated Tx FIFOs mode

[    2.936299] WARN::dwc_otg_hcd_init:1074: FIQ DMA bounce buffers: virt = 9bd04000 dma = 0xdbd04000 len=9024
[    2.936330] FIQ FSM acceleration enabled for :
               Non-periodic Split Transactions
               Periodic Split Transactions
               High-Speed Isochronous Endpoints
               Interrupt/Control Split Transaction hack enabled

[    2.936420] WARN::hcd_init_fiq:457: FIQ on core 1

[    2.936437] WARN::hcd_init_fiq:458: FIQ ASM at 807c9984 length 36

[    2.936455] WARN::hcd_init_fiq:497: MPHI regs_base at a0810000
[    2.936621] Init: Port Power? op_state=1
[    2.936633] Init: Power Port (0)
[    3.034603] mmc1: queuing unknown CIS tuple 0x80 (2 bytes)
[    3.036284] mmc1: queuing unknown CIS tuple 0x80 (3 bytes)
[    3.037965] mmc1: queuing unknown CIS tuple 0x80 (3 bytes)
[    3.040974] mmc1: queuing unknown CIS tuple 0x80 (7 bytes)
[    3.066357] Indeed it is in host mode hprt0 = 00021501
[    3.193653] mmc0: host does not support reading read-only switch, assuming write-enable
[    3.274732] Indeed it is in host mode hprt0 = 00001101
[    5.385760] mcp25xxfd: loading out-of-tree module taints kernel.
[    6.697899] vc_sm_cma: module is from the staging directory, the quality is unknown, you have been warned.
[    6.806256] snd_bcm2835: module is from the staging directory, the quality is unknown, you have been warned.
[    6.809119] bcm2835_mmal_vchiq: module is from the staging directory, the quality is unknown, you have been warned.
[    6.826432] bcm2835_v4l2: module is from the staging directory, the quality is unknown, you have been warned.
[    6.832909] bcm2835_codec: module is from the staging directory, the quality is unknown, you have been warned.
[    6.834240] bcm2835_isp: module is from the staging directory, the quality is unknown, you have been warned.
[    7.376744] brcmfmac: brcmf_fw_alloc_request: using brcm/brcmfmac43430-sdio for chip BCM43430/1
[    7.613842] brcmfmac: brcmf_fw_alloc_request: using brcm/brcmfmac43430-sdio for chip BCM43430/1
[    7.617035] brcmfmac: brcmf_fw_alloc_request: using brcm/brcmfmac43430-sdio for chip BCM43430/1
[    7.617165] brcmfmac: brcmf_c_process_clm_blob: no clm_blob available (err=-2), device may have limited channels available
[    7.618130] brcmfmac: brcmf_c_preinit_dcmds: Firmware: BCM43430/1 wl0: Oct 22 2019 01:59:28 version 7.45.98.94 (r723000 CY) FWID 01-3b33decd
[    8.047083] mcp251x spi1.1: Cannot initialize MCP2515. Wrong wiring?
[    8.047264] mcp251x spi1.1: Probe failed, err=19
[   14.903746] ICMPv6: process `dhcpcd' is using deprecated sysctl (syscall) net.ipv6.neigh.eth0.retrans_time - use net.ipv6.neigh.eth0.retrans_time_ms instead

ad 2. dtoverlay=t-sp1-can1,oscillator=16000000,interrupt=6

         cat /proc/interrupts ---->
          CPU0       CPU1       CPU2       CPU3       
 50:        316          0          0          0  ARMCTRL-level   1 Edge      3f00b880.mailbox
 51:         74          0          0          0  ARMCTRL-level   2 Edge      VCHIQ doorbell
 73:          0          0          0          0  ARMCTRL-level  48 Edge      bcm2708_fb DMA
 75:        213          0          0          0  ARMCTRL-level  50 Edge      DMA IRQ
 77:      10078          0          0          0  ARMCTRL-level  52 Edge      DMA IRQ
 78:          0          0          0          0  ARMCTRL-level  53 Edge      DMA IRQ
 81:          0          0          0          0  ARMCTRL-level  56 Edge      DMA IRQ
 86:          0          0          0          0  ARMCTRL-level  61 Edge      3f215080.spi
 89:      31337          0          0          0  ARMCTRL-level  64 Edge      dwc_otg, dwc_otg_pcd, dwc_otg_hcd:usb1
111:          0          0          0          0  ARMCTRL-level  86 Edge      3f204000.spi
113:        245          0          0          0  ARMCTRL-level  88 Edge      mmc0
114:       4221          0          0          0  ARMCTRL-level  89 Edge      uart-pl011
119:       3265          0          0          0  ARMCTRL-level  94 Edge      mmc1
194:          0          0          0          0  bcm2836-timer   0 Edge      arch_timer
195:       4463       5865       4251       4230  bcm2836-timer   1 Edge      arch_timer
198:          0          0          0          0  bcm2836-pmu   9 Edge      arm-pmu
FIQ:              usb_fiq
IPI0:          0          0          0          0  CPU wakeup interrupts
IPI1:          0          0          0          0  Timer broadcast interrupts
IPI2:        283        303        329        352  Rescheduling interrupts
IPI3:       5106       7433       8897       8871  Function call interrupts
IPI4:          0          0          0          0  CPU stop interrupts
IPI5:        234        166         22         48  IRQ work interrupts
IPI6:          0          0          0          0  completion interrupts
Err:          0

dmesg|-l err,warn
[    0.000000] Kernel parameter elevator= does not have any effect anymore.
               Please use sysfs to set IO scheduler for individual devices.
[    2.737806] Core Release: 2.80a
[    2.737823] Setting default values for core params
[    2.737855] Finished setting default values for core params
[    2.938253] Using Buffer DMA mode
[    2.938266] Periodic Transfer Interrupt Enhancement - disabled
[    2.938278] Multiprocessor Interrupt Enhancement - disabled
[    2.938291] OTG VER PARAM: 0, OTG VER FLAG: 0
[    2.938318] Dedicated Tx FIFOs mode

[    2.938797] WARN::dwc_otg_hcd_init:1074: FIQ DMA bounce buffers: virt = 9bd04000 dma = 0xdbd04000 len=9024
[    2.938830] FIQ FSM acceleration enabled for :
               Non-periodic Split Transactions
               Periodic Split Transactions
               High-Speed Isochronous Endpoints
               Interrupt/Control Split Transaction hack enabled

[    2.938911] WARN::hcd_init_fiq:457: FIQ on core 1

[    2.938927] WARN::hcd_init_fiq:458: FIQ ASM at 807c9984 length 36

[    2.938945] WARN::hcd_init_fiq:497: MPHI regs_base at a0810000
[    2.939110] Init: Port Power? op_state=1
[    2.939122] Init: Power Port (0)
[    3.034920] mmc1: queuing unknown CIS tuple 0x80 (2 bytes)
[    3.036605] mmc1: queuing unknown CIS tuple 0x80 (3 bytes)
[    3.038285] mmc1: queuing unknown CIS tuple 0x80 (3 bytes)
[    3.041295] mmc1: queuing unknown CIS tuple 0x80 (7 bytes)
[    3.068987] Indeed it is in host mode hprt0 = 00021501
[    3.195178] mmc0: host does not support reading read-only switch, assuming write-enable
[    3.274740] Indeed it is in host mode hprt0 = 00001101
[    5.328797] mcp25xxfd: loading out-of-tree module taints kernel.
[    6.624991] vc_sm_cma: module is from the staging directory, the quality is unknown, you have been warned.
[    6.693200] bcm2835_mmal_vchiq: module is from the staging directory, the quality is unknown, you have been warned.
[    6.693207] bcm2835_mmal_vchiq: module is from the staging directory, the quality is unknown, you have been warned.
[    6.694630] snd_bcm2835: module is from the staging directory, the quality is unknown, you have been warned.
[    6.695860] bcm2835_mmal_vchiq: module is from the staging directory, the quality is unknown, you have been warned.
[    6.705363] bcm2835_v4l2: module is from the staging directory, the quality is unknown, you have been warned.
[    6.725878] bcm2835_isp: module is from the staging directory, the quality is unknown, you have been warned.
[    6.730218] bcm2835_codec: module is from the staging directory, the quality is unknown, you have been warned.
[    7.483022] brcmfmac: brcmf_fw_alloc_request: using brcm/brcmfmac43430-sdio for chip BCM43430/1
[    7.725758] brcmfmac: brcmf_fw_alloc_request: using brcm/brcmfmac43430-sdio for chip BCM43430/1
[    7.726869] brcmfmac: brcmf_fw_alloc_request: using brcm/brcmfmac43430-sdio for chip BCM43430/1
[    7.727009] brcmfmac: brcmf_c_process_clm_blob: no clm_blob available (err=-2), device may have limited channels available
[    7.728053] brcmfmac: brcmf_c_preinit_dcmds: Firmware: BCM43430/1 wl0: Oct 22 2019 01:59:28 version 7.45.98.94 (r723000 CY) FWID 01-3b33decd
[    7.996165] mcp251x spi1.0: Cannot initialize MCP2515. Wrong wiring?
[    7.996354] mcp251x spi1.0: Probe failed, err=19
[   15.266635] ICMPv6: process `dhcpcd' is using deprecated sysctl (syscall) net.ipv6.neigh.eth0.retrans_time - use net.ipv6.neigh.eth0.retrans_time_ms instead

ad 2. with both overlays:

        cat /proc/interrupts ---->
          CPU0       CPU1       CPU2       CPU3       
 50:        293            0          0          0  ARMCTRL-level   1 Edge      3f00b880.mailbox
 51:           81            0          0          0  ARMCTRL-level   2 Edge      VCHIQ doorbell
 73:             0            0          0          0  ARMCTRL-level  48 Edge      bcm2708_fb DMA
 75:          213            0          0          0  ARMCTRL-level  50 Edge      DMA IRQ
 77:      10142            0          0          0  ARMCTRL-level  52 Edge      DMA IRQ
 78:              0             0          0          0  ARMCTRL-level  53 Edge      DMA IRQ
 81:              0             0          0          0  ARMCTRL-level  56 Edge      DMA IRQ
 86:              0             0          0          0  ARMCTRL-level  61 Edge      3f215080.spi
 89:      36424             0          0          0  ARMCTRL-level  64 Edge      dwc_otg, dwc_otg_pcd, dwc_otg_hcd:usb1
111:             0             0          0          0  ARMCTRL-level  86 Edge      3f204000.spi
113:         318             0          0          0  ARMCTRL-level  88 Edge      mmc0
114:       4221             0          0          0  ARMCTRL-level  89 Edge      uart-pl011
119:       3934             0          0          0  ARMCTRL-level  94 Edge      mmc1
194:             0             0          0          0  bcm2836-timer   0 Edge      arch_timer
195:       4833       4901       5447       6076  bcm2836-timer   1 Edge      arch_timer
198:             0            0          0          0  bcm2836-pmu   9 Edge      arm-pmu
199:             0            0          0          0  pinctrl-bcm2835   6 Edge      spi1.1
200:      60961          0          0          0  pinctrl-bcm2835   5 Edge      spi1.0
FIQ:              usb_fiq
IPI0:          0          0          0          0  CPU wakeup interrupts
IPI1:          0          0          0          0  Timer broadcast interrupts
IPI2:        296        375        357        332  Rescheduling interrupts
IPI3:      45471      70209      54030      41473  Function call interrupts
IPI4:          0          0          0          0  CPU stop interrupts
IPI5:        126        200        204         91  IRQ work interrupts
IPI6:          0          0          0          0  completion interrupts
Err:          0

dmesg -l err,warn ---->
[    0.000000] Kernel parameter elevator= does not have any effect anymore.
               Please use sysfs to set IO scheduler for individual devices.
[    2.735651] Core Release: 2.80a
[    2.735668] Setting default values for core params
[    2.735699] Finished setting default values for core params
[    2.936093] Using Buffer DMA mode
[    2.936106] Periodic Transfer Interrupt Enhancement - disabled
[    2.936118] Multiprocessor Interrupt Enhancement - disabled
[    2.936131] OTG VER PARAM: 0, OTG VER FLAG: 0
[    2.936159] Dedicated Tx FIFOs mode

[    2.936633] WARN::dwc_otg_hcd_init:1074: FIQ DMA bounce buffers: virt = 9bd04000 dma = 0xdbd04000 len=9024
[    2.936666] FIQ FSM acceleration enabled for :
               Non-periodic Split Transactions
               Periodic Split Transactions
               High-Speed Isochronous Endpoints
               Interrupt/Control Split Transaction hack enabled

[    2.936750] WARN::hcd_init_fiq:457: FIQ on core 1

[    2.936766] WARN::hcd_init_fiq:458: FIQ ASM at 807c9984 length 36

[    2.936784] WARN::hcd_init_fiq:497: MPHI regs_base at a0810000
[    2.936949] Init: Port Power? op_state=1
[    2.936961] Init: Power Port (0)
[    3.034826] mmc1: queuing unknown CIS tuple 0x80 (2 bytes)
[    3.036507] mmc1: queuing unknown CIS tuple 0x80 (3 bytes)
[    3.038187] mmc1: queuing unknown CIS tuple 0x80 (3 bytes)
[    3.041196] mmc1: queuing unknown CIS tuple 0x80 (7 bytes)
[    3.068911] Indeed it is in host mode hprt0 = 00021501
[    3.196028] mmc0: host does not support reading read-only switch, assuming write-enable
[    3.275109] Indeed it is in host mode hprt0 = 00001101
[    5.191861] mcp25xxfd: loading out-of-tree module taints kernel.
[    6.522456] vc_sm_cma: module is from the staging directory, the quality is unknown, you have been warned.
[    6.647748] snd_bcm2835: module is from the staging directory, the quality is unknown, you have been warned.
[    6.655471] bcm2835_mmal_vchiq: module is from the staging directory, the quality is unknown, you have been warned.
[    6.656613] bcm2835_mmal_vchiq: module is from the staging directory, the quality is unknown, you have been warned.
[    6.663784] bcm2835_v4l2: module is from the staging directory, the quality is unknown, you have been warned.
[    6.690741] bcm2835_isp: module is from the staging directory, the quality is unknown, you have been warned.
[    6.864758] bcm2835_codec: module is from the staging directory, the quality is unknown, you have been warned.
[    7.294801] brcmfmac: brcmf_fw_alloc_request: using brcm/brcmfmac43430-sdio for chip BCM43430/1
[    7.535991] brcmfmac: brcmf_fw_alloc_request: using brcm/brcmfmac43430-sdio for chip BCM43430/1
[    7.571407] brcmfmac: brcmf_fw_alloc_request: using brcm/brcmfmac43430-sdio for chip BCM43430/1
[    7.571543] brcmfmac: brcmf_c_process_clm_blob: no clm_blob available (err=-2), device may have limited channels available
[    7.572570] brcmfmac: brcmf_c_preinit_dcmds: Firmware: BCM43430/1 wl0: Oct 22 2019 01:59:28 version 7.45.98.94 (r723000 CY) FWID 01-3b33decd
[   14.078073] ICMPv6: process `dhcpcd' is using deprecated sysctl (syscall) net.ipv6.neigh.eth0.retrans_time - use net.ipv6.neigh.eth0.retrans_time_ms instead

Strange isnt it?

More playing around with the overlays (maybe this information helps)

dtoverlay=t-sp1-can1,oscillator=16000000,interrupt=5
dtoverlay=t-sp1-can0,oscillator=16000000,interrupt=6

FAILED Failed to start Raise network interfaces
dmesg|grep spi----->[    7.975624] mcp251x spi1.0 can0: MCP2515 successfully initialized.
                                 [    7.987713] mcp251x spi1.1 can1: MCP2515 successfully initialized.
                                 [    8.019905] mcp251x spi1.0 rename4: renamed from can0

dtoverlay=t-sp1-can0,oscillator=16000000,interrupt=6
dtoverlay=t-sp1-can1,oscillator=16000000,interrupt=5
dmesg|grep spi---->[    8.062882] mcp251x spi1.1 can0: MCP2515 successfully initialized.
                                [    8.075177] mcp251x spi1.0 can1: MCP2515 successfully initialized.

but can0 and can1 not OK

@pelwell
Copy link
Contributor

pelwell commented Feb 7, 2023

I haven't had a chance to digest all of that information yet, but are those results repeatable?

@martburg
Copy link
Author

martburg commented Feb 7, 2023

Yes they are

@pelwell
Copy link
Contributor

pelwell commented Feb 7, 2023

  1. https://www.waveshare.com/wiki/2-CH_CAN_HAT appears to be designed for SPI0. How have you changed it to use SPI1? Are you sure all the signals are connected correctly?

  2. I think these results are significant:

199:        0       0       0       0  pinctrl-bcm2835   6 Edge      spi1.1
200:    60961       0       0       0  pinctrl-bcm2835   5 Edge      spi1.0

For some reason, spi1.1 is not getting any interrupts. Does this change if you momentarily ground GPIO6/pin 29?

  1. The interrupts are declared as falling-edge (interrupt flags 2 is IRQ_TYPE_EDGE_FALLING). Are those pins reading back as high normally (raspi-gpio get 5,6)

By the way, the overlay fragments disabling spidev0 and spidev1 are misguided - they are actually disabling spidev0.0 and spidev0.1, and by enabling SPI1 with dtoverlay=spi1-2cs,cs0_spidev=off,cs1_spidev=off you have already disabled the spidevs.

@martburg
Copy link
Author

martburg commented Feb 8, 2023

ad 5. using J9 on the Board
rechecked the umptiest time
there is another detail to add
up until now I feed the same signal to both can chanels by connecting can0 hi to can1 hi can0 lo to can1 lo using the
jumpers on the board
cutting those connetctions I realised ------ wait for it ----
physically feeding the can0 jumper with the signal candump can1 shows the input

ad 6-7. Time to get the osziloscope: int5 (asociated with can0) gets the interupt
verrified by running raspi-gpio get 5,6 in a loop confirms that GPIO 5 is toggeling
wheras GPIO 6 is LOW

I suspected that the fragments were misguided but my understanding of the dt was/is not good enough to know what I'm doing

just for the record 1:
candump can1 out printed
can1 49D [6] 70 16 00 00 00 00
can1 000 [8] FD 02 FD FF FF 7E 57 6F
can1 12D [7] remote request
can1 1EF9FE66 [8] remote request

just for the record2:
this is the dmesg|grep spi output if the spi-2s overlay is not loaded:
[ 6.899760] spi-bcm2835aux 3f215080.spi: cs1 >= max 1
[ 6.899805] spi_master spi1: spi_device register error /soc/spi@7e215080/mcp2515@1
[ 6.899836] spi_master spi1: Failed to create SPI device for /soc/spi@7e215080/mcp2515@1
[ 6.899904] spi spi1.0: Native CS is not supported - please configure cs-gpio in device-tree
[ 8.133857] mcp251x spi1.0: Native CS is not supported - please configure cs-gpio in device-tree
[ 9.159534] mcp251x spi1.0: MCP251x didn't enter in conf mode after reset
[ 9.159724] mcp251x spi1.0: Probe failed, err=110
[ 9.159808] mcp251x: probe of spi1.0 failed with error -110

@pelwell
Copy link
Contributor

pelwell commented Feb 8, 2023

physically feeding the can0 jumper with the signal candump can1 shows the input

It's important to distinguish between the bus numbers on the board and the labels that Linux gives the network interfaces. To make sure I've understood you correctly, please do the following:

  1. Attach the CAN devices to the CAN0 connector, boot, and run the following: 
    $ candump can0
$ candump can1
$ dmesg | grep "MCP2515 successfully initialized"
$ grep spi1 /proc/interrupts
  2. As 8, but attach the devices to CAN1 instead.

@martburg
Copy link
Author

martburg commented Feb 8, 2023

ad 8. CAN on CAN0 connector

candump can0 no output
candump can1 output ok

dmesg|grep spi
[ 7.951712] mcp251x spi1.1 can0: MCP2515 successfully initialized.
[ 7.964464] mcp251x spi1.0 can1: MCP2515 successfully initialized.

grep spi1 /proc/interrupts
199: 0 0 0 0 pinctrl-bcm2835 6 Edge spi1.1
200: 121014 0 0 0 pinctrl-bcm2835 5 Edge spi1.0

ad 9. CAN on CAN1 connector
candump can0 no output
candump can1 no output

dmesg|grep spi
[ 8.040167] mcp251x spi1.1 can0: MCP2515 successfully initialized.
[ 8.054469] mcp251x spi1.0 can1: MCP2515 successfully initialized.

grep spi1 /proc/interrupts
199: 0 0 0 0 pinctrl-bcm2835 6 Edge spi1.1
200: 16 0 0 0 pinctrl-bcm2835 5 Edge spi1.0

@pelwell
Copy link
Contributor

pelwell commented Feb 9, 2023

In both cases the driver can communicate with both instances of the MCP2515. If one of them was responding on both chip selects then the results would have been the same for 8 & 9, but they weren't. From that I conclude that the SPI communication is working.

There's still a slight question mark over the status of the sp1.1 interrupt on GPIO 6. Do you ever see it high? Could a stuck interrupt line appear to show a working controller but with no devices?

Whether the interrupt line is the cause of the problem, or whether it's just a symptom of fault on the CAN side of the MCP2515, it definitely smells like a hardware problem on the CAN board. Have you tried using it on SPI0 (with J9 moved)?

@martburg
Copy link
Author

martburg commented Feb 9, 2023

I repeated the whole procedure with an identical board. Same results
I have shipped an other rpi where the mcp2515 was connected to spi0 but I'm not shure if it was the identical board but that could have been an other version.

I'll try to move J9 to SPI0 or should I try to move int6 to another pin?

@pelwell
Copy link
Contributor

pelwell commented Feb 9, 2023

Either, or both - the aim is to try different combinations to help to pinpoint the problem.

@martburg
Copy link
Author

martburg commented Feb 9, 2023

I put our converation on stackoverflow
https://raspberrypi.stackexchange.com/questions/141017/2-ch-can-hat-mcp2515-on-spi1-rpi-3-b
and opened a ticket on the waveshare support

Now I'll will head over to the lab and try some combinations.

@pelwell : Thank you for the support!! I hope we can corner this bug.

@martburg
Copy link
Author

martburg commented Feb 9, 2023

connected the 2-ch-can-hat to SPI0

INT1 --- Pin 22 GPIO 25
INT0 --- Pin 18 GPIO 23
CS1 ---- Pin 26 GPIO 7 (SPI0 CE1)
CS0 ---- Pin 24 GPIO 8 (SPI0 CE0)
SCK ---- Pin 23 GPIO 11 (SPI0 SCLK)
MOSI -- Pin 19 GPIO 10 (SPI0 MOSI)
MISO -- Pin 21 GPIO 9 (SPI0 MISO)
GND --- Pin 6 GND
5V ----- Pin 1 5V

/boot/config.txt:
dtoverlay=mcp2515-can0,oscillator=16000000,interrupt=23
dtoverlay=mcp2515-can1,oscillator=16000000,interrupt=25

mcp2515-can0.dtbo compiles from:

 /*
  * Device tree overlay for mcp251x/can0 on spi0.0
 */

/dts-v1/;
/plugin/;

/ {
    compatible = "brcm,bcm2835";
    /* disable spi-dev for spi0.0 */
    fragment@0 {
        target = <&spi0>;
        __overlay__ {
            status = "okay";
        };
    };

    fragment@1 {
    target = <&spidev0>;
    __overlay__ {
        status = "disabled";
    };
    };

    /* the interrupt pin of the can-controller */
    fragment@2 {
        target = <&gpio>;
        __overlay__ {
            can0_pins: can0_pins {
                brcm,pins = <25>;
                brcm,function = <0>; /* input */
            };
        };
    };

    /* the clock/oscillator of the can-controller */
    fragment@3 {
        target-path = "/";
        __overlay__ {
            /* external oscillator of mcp2515 on SPI0.0 */
                can0_osc: can0_osc {
                compatible = "fixed-clock";
                #clock-cells = <0>;
                clock-frequency  = <16000000>;
            };
        };
    };

    /* the spi config of the can-controller itself binding everything together */
    fragment@4 {
        target = <&spi0>;
        __overlay__ {
            /* needed to avoid dtc warning */
            #address-cells = <1>;
            #size-cells = <0>;
            can0: mcp2515@0 {
                reg = <0>;
                compatible = "microchip,mcp2515";
                pinctrl-names = "default";
                pinctrl-0 = <&can0_pins>;
                spi-max-frequency = <10000000>;
                interrupt-parent = <&gpio>;
                interrupts = <25 8>; /* IRQ_TYPE_LEVEL_LOW */
                clocks = <&can0_osc>;
            };
        };
    };
    __overrides__ {
        oscillator = <&can0_osc>,"clock-frequency:0";
        spimaxfrequency = <&can0>,"spi-max-frequency:0";
        interrupt = <&can0_pins>,"brcm,pins:0",<&can0>,"interrupts:0";
    };
};`

mcp2515-can1.dtbo compiles from

 /*
 * Device tree overlay for mcp251x/can1 on spi0.1 edited by petit_miner
 */

/dts-v1/;
/plugin/;

/ {
    compatible = "brcm,bcm2835";
    /* disable spi-dev for spi0.1 */
    fragment@0 {
        target = <&spi0>;
        __overlay__ {
            status = "okay";
        };
    };

    fragment@1 {
    target = <&spidev1>;
    __overlay__ {
        status = "disabled";
    };
    };

    /* the interrupt pin of the can-controller */
    fragment@2 {
        target = <&gpio>;
        __overlay__ {
            can1_pins: can1_pins {
                brcm,pins = <25>;
                brcm,function = <0>; /* input */
            };
        };
    };

    /* the clock/oscillator of the can-controller */
    fragment@3 {
        target-path = "/";
        __overlay__ {
            /* external oscillator of mcp2515 on spi0.1 */
            can1_osc: can1_osc {
                compatible = "fixed-clock";
                #clock-cells = <0>;
                clock-frequency  = <16000000>;
            };
        };
    };

    /* the spi config of the can-controller itself binding everything together */
    fragment@4 {
        target = <&spi0>;
        __overlay__ {
            /* needed to avoid dtc warning */
            #address-cells = <1>;
            #size-cells = <0>;
            can1: mcp2515@1 {
                reg = <1>;
                compatible = "microchip,mcp2515";
                pinctrl-names = "default";
                pinctrl-0 = <&can1_pins>;
                spi-max-frequency = <10000000>;
                interrupt-parent = <&gpio>;
                interrupts = <25 8>; /* IRQ_TYPE_LEVEL_LOW */
                clocks = <&can1_osc>;
            };
        };
    };
    __overrides__ {
        oscillator = <&can1_osc>,"clock-frequency:0";
        spimaxfrequency = <&can1>,"spi-max-frequency:0";
        interrupt = <&can1_pins>,"brcm,pins:0",<&can1>,"interrupts:0";
    };
};

dmesg | grep spi

[ 8.132198] mcp251x spi0.1: prop pinctrl-0 index 0 invalid phandle
[ 8.132254] mcp251x: probe of spi0.1 failed with error -22
[ 8.132359] mcp251x spi0.0: prop pinctrl-0 index 0 invalid phandle
[ 8.132384] mcp251x: probe of spi0.0 failed with error -22

So I got a fresh copy of

mcp2515-can0.dtbo and mcp2515-can1.dtbo

from https://github.com/raspberrypi/firmware/tree/master/boot/overlays

and got the same

[ 7.856228] mcp251x spi0.1: prop pinctrl-0 index 0 invalid phandle
[ 7.856278] mcp251x: probe of spi0.1 failed with error -22
[ 7.856354] mcp251x spi0.0: prop pinctrl-0 index 0 invalid phandle
[ 7.856379] mcp251x: probe of spi0.0 failed with error -22

@pelwell
Copy link
Contributor

pelwell commented Feb 9, 2023

How old is your firmware (vcgencmd version)?

@martburg
Copy link
Author

martburg commented Feb 9, 2023

System

Raspberry Pi reference 2021-01-11
Generated using pi-gen, https://github.com/RPi-Distro/pi-gen, 21090519d85bdaa1615d5d5057d37b09368ea5d2, stage2

Oct 29 2021 10:49:08
Copyright (c) 2012 Broadcom
version b8a114e5a9877e91ca8f26d1a5ce904b2ad3cf13 (clean) (release) (start)

Linux raspberrypi 5.10.63-v7+ #1459 SMP Wed Oct 6 16:41:10 BST 2021 armv7l GNU/Linux

@pelwell
Copy link
Contributor

pelwell commented Feb 9, 2023

Don't worry, I've found it - Oct 29 2021 10:49:08. I've just tried those overlays here, and although the mcp2515 driver can't find any hardware (obviously), there is no problem with the overlay application:

~$ dmesg | grep mcp
[   10.963234] mcp251x spi0.1: MCP251x didn't enter in conf mode after reset
[   10.964550] mcp251x spi0.1: Probe failed, err=110
[   10.964577] mcp251x: probe of spi0.1 failed with error -110
[   11.982433] mcp251x spi0.0: MCP251x didn't enter in conf mode after reset
[   11.982626] mcp251x spi0.0: Probe failed, err=110
[   11.982637] mcp251x: probe of spi0.0 failed with error -110

Try grabbing the overlays from the same release as the firmware: https://github.com/raspberrypi/rpi-firmware/tree/d5edc6af1ef48f97b525da88ff6c510c2d4231c3/overlays

@martburg
Copy link
Author

martburg commented Feb 9, 2023

Grabed it loaded it ---
pi@raspberrypi:~ $ dmesg|grep spi
[ 7.927692] mcp251x spi0.1: prop pinctrl-0 index 0 invalid phandle
[ 7.927738] mcp251x: probe of spi0.1 failed with error -22
[ 7.927815] mcp251x spi0.0: prop pinctrl-0 index 0 invalid phandle
[ 7.927837] mcp251x: probe of spi0.0 failed with error -22
pi@raspberrypi:~ $ dmesg|grep mcp
[ 5.271546] mcp25xxfd: loading out-of-tree module taints kernel.
[ 7.927692] mcp251x spi0.1: prop pinctrl-0 index 0 invalid phandle
[ 7.927738] mcp251x: probe of spi0.1 failed with error -22
[ 7.927815] mcp251x spi0.0: prop pinctrl-0 index 0 invalid phandle
[ 7.927837] mcp251x: probe of spi0.0 failed with error -22
pi@raspberrypi:~ $

@martburg
Copy link
Author

New System

Raspberry Pi reference 2022-09-22
Generated using pi-gen, https://github.com/RPi-Distro/pi-gen, 005a8c73b05a2cab394073150208bf4f069e861a, stage2

Copyright (c) 2012 Broadcom
version 102f1e848393c2112206fadffaaf86db04e98326 (clean) (release) (start)

Linux raspberrypi 5.15.61-v8+ #1579 SMP PREEMPT Fri Aug 26 11:16:44 BST 2022 aarch64 GNU/Linux

/boot/config.txt:

dtoverlay=mcp2515-can0,oscillator=16000000,interrupt=23
dtoverlay=mcp2515-can1,oscillator=16000000,interrupt=25

no Board attached -->

[ 11.540216] mcp251x spi0.1: MCP251x didn't enter in conf mode after reset
[ 11.541770] mcp251x spi0.1: Probe failed, err=110
[ 11.541833] mcp251x: probe of spi0.1 failed with error -110
[ 12.566084] mcp251x spi0.0: MCP251x didn't enter in conf mode after reset
[ 12.566333] mcp251x spi0.0: Probe failed, err=110
[ 12.566350] mcp251x: probe of spi0.0 failed with error -110

Board attached to spi0 -->

[ 10.532899] mcp251x spi0.1 can0: MCP2515 successfully initialized.
[ 10.551022] mcp251x spi0.0 can1: MCP2515 successfully initialized.

Still the mixup CAN0 and CAN1 but both channels working on SPI0

Next Step ---> move to SPI1

I'll keep you posted

@martburg
Copy link
Author

one Overlay:
dt-overlay=t-spi1-can0,oscillator=16000000,interrupt=5

[ 11.608247] mcp251x spi1.0: Cannot initialize MCP2515. Wrong wiring?
[ 11.608751] mcp251x spi1.0: Probe failed, err=19

dt-overlay=t-spi1-can1,oscillator=16000000,interrupt=6

[ 10.718563] mcp251x spi1.1: Cannot initialize MCP2515. Wrong wiring?
[ 10.728394] mcp251x spi1.1: Probe failed, err=19

both Overlays
dt-overlay=t-spi1-can0,oscillator=16000000,interrupt=5
dt-overlay=t-spi1-can1,oscillator=16000000,interrupt=6

[ 10.797097] mcp251x spi1.1 can0: MCP2515 successfully initialized.
[ 10.873963] mcp251x spi1.0 can1: MCP2515 successfully initialized.

candump can0 (asociated with spi1.1) not OK
candump can1 (asociated with spi 1.0) OK

cat /proc/interrupts
184: 0 0 0 0 pinctrl-bcm2835 6 Edge spi1.1
185: 68932 76684 71270 76552 pinctrl-bcm2835 5 Edge spi1.0

GPIO 5 (int0) HIGH and gets Interrupts
GPIO 6 (int1) is LOW

looks like we are back on square one - but we verrified the board is working on spi0 (with crosover between can0 and can1)

and the new system makes a more stable impression

@martburg
Copy link
Author

martburg commented Feb 10, 2023 •
edited
Loading

changing the order of the loading of the overlays took care of the crosover
both of the board beeing on spi0 or spi1

on spi0 both channels ar working
on spi1 only one is working

so we could change the issue to

mcp2515 on spi1 only one can channel working

@pelwell
Copy link
Contributor

pelwell commented Feb 10, 2023

I think this is a hardware problem, one that few people have encountered because most would use it on SPI0.

@martburg
Copy link
Author

Shure?
As I was using J9 on the waveshare board both for the connection to spi0 and spi1 for me the board looks OK.
but as I need only can0 on spi1 for this project --- I vote for closing this issue, if somebody else or I at a later time need both channels working we already have the title for that issue.

One last thing tomorow I'll try to change from GPIO 6 maybe this is the issue

@martburg
Copy link
Author

Moving the interrupt vrom pin 31 (GPIO 6) to pin 33 (GPIO 13) solved the problem.

So either it is just my rpi where GPIO 6 in this combination is faulty or ....

Any way -- hardware

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@martburg @pelwell