Error implementing Sony IMX219 with 8MNano EVK

cancel
Showing results for 
Search instead for 
Did you mean: 

Error implementing Sony IMX219 with 8MNano EVK

156 Views
artur_malinouski
Contributor I

Hi all! First post!

I'm trying to get the Sony IMX219 sensor to work with the iMX 8M Nano EVK.

I used bitbake linux-imx -c menuconfig to set the IMX219 driver to "built-in," disabled the OV5640 driver and built the latest kernel v5.10.9-1.0.0. I then used that kernel image in a fresh imx-image-multimedia.

For some reason, the auto-generated .dtb file did not have a node for the IMX219, yet still had one for the OV5640. I adapted the OV5640 node for my purposes by changing the device name, pointing to the correct i2c bus address, using named &links to connect the camera endpoint to the csi endpoint and adding a "link-frequencies" property as I've seen others do on these forums. Here's what the modified parts of my DTB look like:

csi@32e30000 {
compatible = "fsl,imx8mn-mipi-csi";
reg = <0x0 0x32e30000 0x0 0x10000>;
interrupts = <0x0 0x11 0x4>;
clock-frequency = <0x13d92d40>;
clocks = <0x2 0x81 0x2 0xb8 0x2 0xb9>;
clock-names = "mipi_clk", "disp_axi", "disp_apb";
assigned-clocks = <0x2 0x81>;
assigned-clock-parents = <0x2 0x41>;
assigned-clock-rates = <0x13d92d40>;
bus-width = <0x4>;
csi-gpr = <0x14>;
power-domains = <0x15>;
resets = <0x16>;
status = "okay";
#address-cells = <0x1>;
#size-cells = <0x0>;

port@0 {
reg = <0x0>;

csi0_1: endpoint {
remote-endpoint = <&imx219_1>;
data-lanes = <0x2>;
csis-hs-settle = <0xd>;
csis-clk-settle = <0x2>;
csis-wclk;
};
};
};

.........

imx219@10 {
compatible = "sony,imx219";
reg = <0x10>;
pinctrl-names = "default";
pinctrl-0 = <0x3a 0x3b>;
clocks = <0x2 0x7b>;
clock-names = "xclk";
assigned-clocks = <0x2 0x7b>;
assigned-clock-parents = <0x2 0x2>;
assigned-clock-rates = <0x16e3600>;
csi_id = <0x0>;
powerdown-gpios = <0x2e 0x7 0x0>;
reset-gpios = <0x2e 0x6 0x1>;
mclk = <0x16e3600>;
mclk_source = <0x0>;
mipi_csi;
status = "okay";

port {

imx219_1: endpoint {
remote-endpoint = <&csi0_1>;
data-lanes = <0x1 0x2>;
clocks-lanes = <0x0>;
link-frequencies = /bits/ 64 <456000000>;
};
};
};

I recompiled and replaced the DTB on the boot partition of my SD card. The image boots successfully, and I can see UU on register 10 on I2C2 with i2cdetect. However, /dev/video0 does not register. Looking at boot logs, I see the following kernel oops error:

[ 5.573945] imx219 2-0010: supply VDIG not found, using dummy regulator
[ 5.583143] bd718xx-clk bd71847-clk.2.auto: No parent clk found
socket.
[ 5.592300] caam 30900000.crypto: job rings = 2, qi = 0
[ 5.593749] bd718xx-clk: probe of bd71847-clk.2.auto failed with error -22
[ 5.608096] imx219 2-0010: supply VDDL not found, using dummy regulator
[ 5.648374] mx8-img-md: Registered sensor subdevice: imx219 2-0010 (1)
[ 5.657399] ------------[ cut here ]------------
[ 5.662177] WARNING: CPU: 0 PID: 215 at drivers/media/mc/mc-entity.c:665 media_create_pad_link+0x190/0x210
[ 5.671838] Modules linked in: clk_bd718x7 imx219(+) caam secvio imx8_media_dev(C) error gpio_ir_recv rc_core
[ 5.681784] CPU: 0 PID: 215 Comm: systemd-udevd Tainted: G C 5.10.9-1.0.0+g32513c25d8c7 #1
[ 5.691349] Hardware name: NXP i.MX8MNano DDR4 EVK board (DT)
[ 5.697102] pstate: 40000005 (nZcv daif -PAN -UAO -TCO BTYPE=--)
[ 5.703110] pc : media_create_pad_link+0x190/0x210
[ 5.707905] lr : subdev_notifier_complete+0x128/0x850 [imx8_media_dev]
[ 5.714432] sp : ffff8000121f3760
[ 5.717745] x29: ffff8000121f3760 x28: ffff00000593c884
[ 5.723058] x27: ffff00000593c6e0 x26: ffff800009136530
[ 5.728370] x25: ffff00000593c080 x24: ffff00000593c0a4
[ 5.733683] x23: ffff00000593c8e0 x22: ffff00000593c0c4
[ 5.738995] x21: 000000000000000c x20: 0000000000000000
[ 5.744307] x19: 0000000000000000 x18: 0000000000000000
[ 5.749619] x17: 0000000000000000 x16: 0000000000000000
[ 5.754930] x15: 00008c25633332d6 x14: 0000000000000140
[ 5.760242] x13: 0000000000000140 x12: 0000000000000000
[ 5.765554] x11: 0000000000000007 x10: 0101010101010101
[ 5.770865] x9 : 0000000000000004 x8 : 0000000000000004
[ 5.776177] x7 : 0000000000000008 x6 : dead000000000100
[ 5.781489] x5 : ffff000005f0d190 x4 : 0000000000000001
[ 5.786801] x3 : 0000000000000000 x2 : 0000000000000000
[ 5.792112] x1 : 000000000000000c x0 : 0000000000000000
[ 5.797424] Call trace:
[ 5.799873] media_create_pad_link+0x190/0x210
[ 5.804316] v4l2_async_notifier_try_complete.part.0+0x58/0x64
[ 5.810148] v4l2_async_register_subdev+0x100/0x1c0
[ 5.815025] v4l2_async_register_subdev_sensor_common+0x6c/0xf0
[ 5.820948] imx219_probe+0x5c8/0x7ac [imx219]
[ 5.825392] i2c_device_probe+0x118/0x300
[ 5.829403] really_probe+0xe4/0x490
[ 5.832977] driver_probe_device+0x58/0xc0
[ 5.837073] device_driver_attach+0xc0/0xd0
[ 5.841254] __driver_attach+0x84/0x124
[ 5.845090] bus_for_each_dev+0x70/0xc0
[ 5.848924] driver_attach+0x24/0x30
[ 5.852498] bus_add_driver+0x104/0x1ec
[ 5.856331] driver_register+0x78/0x130
[ 5.860165] i2c_register_driver+0x50/0xd0
[ 5.864264] imx219_i2c_driver_init+0x24/0x1000 [imx219]
[ 5.869576] do_one_initcall+0x54/0x1bc
[ 5.873411] do_init_module+0x54/0x250
[ 5.877158] load_module+0x1ea8/0x25c0
[ 5.880905] __do_sys_finit_module+0xb8/0x100
[ 5.885259] __arm64_sys_finit_module+0x24/0x30
[ 5.889790] el0_svc_common.constprop.0+0x78/0x1a0
[ 5.894578] do_el0_svc+0x24/0x90
[ 5.897893] el0_svc+0x14/0x20
[ 5.900946] el0_sync_handler+0x1a4/0x1b0
[ 5.904953] el0_sync+0x174/0x180
[ 5.908266] ---[ end trace 66170a6aebb9ae21 ]---
[ 5.913219] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000020
[ 5.922234] Mem abort info:
[ 5.925143] ESR = 0x96000004
[ 5.928275] EC = 0x25: DABT (current EL), IL = 32 bits
[ 5.933651] SET = 0, FnV = 0
[ 5.936802] EA = 0, S1PTW = 0
[ 5.939962] Data abort info:
[ 5.942967] ISV = 0, ISS = 0x00000004
[ 5.946893] CM = 0, WnR = 0
[ 5.949907] user pgtable: 4k pages, 48-bit VAs, pgdp=0000000045c1f000
[ 5.956414] [0000000000000020] pgd=0000000000000000, p4d=0000000000000000
[ 5.963312] Internal error: Oops: 96000004 [#1] PREEMPT SMP
[ 5.968889] Modules linked in: clk_bd718x7 imx219(+) caam secvio imx8_media_dev(C) error gpio_ir_recv rc_core
[ 5.978832] CPU: 0 PID: 215 Comm: systemd-udevd Tainted: G WC 5.10.9-1.0.0+g32513c25d8c7 #1
[ 5.988394] Hardware name: NXP i.MX8MNano DDR4 EVK board (DT)
[ 5.994140] pstate: 40000005 (nZcv daif -PAN -UAO -TCO BTYPE=--)
[ 6.000149] pc : subdev_notifier_complete+0x5b4/0x850 [imx8_media_dev]
[ 6.006676] lr : subdev_notifier_complete+0x128/0x850 [imx8_media_dev]
[ 6.013200] sp : ffff8000121f3760
[ 6.016512] x29: ffff8000121f3760 x28: ffff00000593c884
[ 6.021824] x27: ffff00000593c6e0 x26: ffff800009136530
[ 6.027137] x25: ffff00000593c080 x24: ffff00000593c0a4
[ 6.032448] x23: ffff00000593c8e0 x22: ffff00000593c0c4
[ 6.037760] x21: 000000000000000c x20: 0000000000000000
[ 6.043072] x19: 0000000000000000 x18: 0000000000000000
[ 6.048383] x17: 0000000000000000 x16: 0000000000000000
[ 6.053695] x15: 00008c25633332d6 x14: 0000000000000140
[ 6.059006] x13: 0000000000000140 x12: 0000000000000000
[ 6.064318] x11: 0000000000000007 x10: 0101010101010101
[ 6.069629] x9 : 0000000000000004 x8 : 0000000000000004
[ 6.074941] x7 : 0000000000000008 x6 : dead000000000100
[ 6.080253] x5 : ffff000005f0d190 x4 : 0000000000000001
[ 6.085565] x3 : 0000000000000000 x2 : 0000000000000000
[ 6.090876] x1 : 000000000000000c x0 : 00000000ffffffea
[ 6.096189] Call trace:
[ 6.098636] subdev_notifier_complete+0x5b4/0x850 [imx8_media_dev]
[ 6.104819] v4l2_async_notifier_try_complete.part.0+0x58/0x64
[ 6.110650] v4l2_async_register_subdev+0x100/0x1c0
[ 6.115528] v4l2_async_register_subdev_sensor_common+0x6c/0xf0
[ 6.121450] imx219_probe+0x5c8/0x7ac [imx219]
[ 6.125893] i2c_device_probe+0x118/0x300
[ 6.129904] really_probe+0xe4/0x490
[ 6.133478] driver_probe_device+0x58/0xc0
[ 6.137573] device_driver_attach+0xc0/0xd0
[ 6.141754] __driver_attach+0x84/0x124
[ 6.145589] bus_for_each_dev+0x70/0xc0
[ 6.149423] driver_attach+0x24/0x30
[ 6.152998] bus_add_driver+0x104/0x1ec
[ 6.156831] driver_register+0x78/0x130
[ 6.160665] i2c_register_driver+0x50/0xd0
[ 6.164762] imx219_i2c_driver_init+0x24/0x1000 [imx219]
[ 6.170074] do_one_initcall+0x54/0x1bc
[ 6.173909] do_init_module+0x54/0x250
[ 6.177657] load_module+0x1ea8/0x25c0
[ 6.181404] __do_sys_finit_module+0xb8/0x100
[ 6.185758] __arm64_sys_finit_module+0x24/0x30
[ 6.190289] el0_svc_common.constprop.0+0x78/0x1a0
[ 6.195080] do_el0_svc+0x24/0x90
[ 6.198394] el0_svc+0x14/0x20
[ 6.201447] el0_sync_handler+0x1a4/0x1b0
[ 6.205454] el0_sync+0x174/0x180
[ 6.208772] Code: d0000000 9114c000 f90047e0 17ffffe5 (f9401262)
[ 6.214866] ---[ end trace 66170a6aebb9ae22 ]---

I don't have the skills to debug this, so I was hoping I could get some help!

@igorpadykov 

0 Kudos
2 Replies

112 Views
artur_malinouski
Contributor I

Hi Igor,

Thanks for your reply. I ran the DDR stress test and it passed OK, see attached log.

Based on the stack trace from the first crash, the error seems to occur in the media_create_pad_link function of the imx8_media_dev driver.

I also scrolled up in the boot log a little further and noticed this prior to the crash:

[    6.436688] mx8-img-md: Registered mxc_isi.0.capture as /dev/video0

[    6.444759] unregister ISI channel: mxc_isi.0

The unregister ISI channel event does not occur when I boot with the ov5640, but does occur with the imx219 and seems to be related to the media_pad_link crash.

What would you recommend next?

0 Kudos

153 Views
igorpadykov
NXP TechSupport
NXP TechSupport

Hi Artur

 

log error "Unable to handle kernel NULL pointer dereference at virtual address 0000000000000020"

may point to ddr errors, so one can try to run ddr test and update image with new ddr settings

following ddr test documentation

https://community.nxp.com/t5/i-MX-Processors-Knowledge-Base/i-MX-8M-Family-DDR-Tool-Release/ta-p/110...

 

Best regards
igor

0 Kudos