Help with kernel panic

[timothybean]

Hello Freescale guys! Can anyone help me with figuring out the resolution to the following kernel panic? This only happens after hours of playing video. Nothing seems to "trigger" it... just happens on its own.

Unable to handle kernel NULL pointer dereference at virtual address 00000000

pgd = 80004000

[00000000] *pgd=00000000

Internal error: Oops: 17 [#1] PREEMPT SMP

Modules linked in:

CPU: 0    Not tainted  (3.0.35-g0642749-dirty #207)

PC is at __wake_up_common+0x28/0x90

LR is at __wake_up+0x3c/0x50

pc : [<80064904>]    lr : [<80064e74>]    psr: a0000093

sp : bfc73c40  ip : 00000000  fp : bfc73c6c

r10: 00000003  r9 : 00000000  r8 : 00000000

r7 : b2e43aac  r6 : 00000000  r5 : 60000013  r4 : b2e43aa4

r3 : 00000000  r2 : 00000001  r1 : 00000003  r0 : b2e43aa4

Flags: NzCv  IRQs off  FIQs on  Mode SVC_32  ISA ARM  Segment kernel

Control: 10c53c7d  Table: 432ec04a  DAC: 00000015

Process ipu1_task (pid: 400, stack limit = 0xbfc722f0)

Stack: (0xbfc73c40 to 0xbfc74000)

3c40: 00000001 00000001 80064e58 b2e43aa4 60000013 00000000 00000001 00000003

3c60: 00200200 00100100 bfc73c94 80064e74 00000000 b2e42600 b2e42600 b2e42600

3c80: 00000001 b2e43a8c bfc73f84 00000000 00000000 803c0e94 bfc73cd4 8008ba54

3ca0: b350d768 8006492c b350d648 00000001 00200200 b350d76c 60000013 00000000

3cc0: 00000001 00000003 00200200 00100100 bfc73cfc 00000001 00000000 00000094

3ce0: 00000000 b2e42600 bfc73f84 00000000 00200200 00100100 00000000 803c1a34

3d00: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000002

3d20: 00000000 bfc72000 00000001 00000001 00000000 00000500 000002d0 3231564e

3d40: 00000270 00000000 00000290 000002d0 32800000 00000000 00000000 00000500

3d60: 000002d0 50424752 00000000 00000270 00000000 00000290 000002d0 18800000

3d80: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000

3da0: 00000000 00000000 00000000 00000000 00000000 00000000 000003f2 b350d600

3dc0: b350d000 0196fd82 00000500 000002d0 3231564e 00000000 00000000 00000290

3de0: 000002d0 32800000 00000000 00000000 00000500 000002d0 50424752 00000000

3e00: 00000000 00000000 00000290 000002d0 18800000 00000000 00000000 00000000

3e20: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000

3e40: 00000000 00000000 00000000 000003f2 b350d600 b350ce00 0196fd81 00000001

3e60: 00000000 00000000 00000000 bfc73f9c bfe8fa00 bfc72000 00000002 bfe8fb94

3e80: ffff8b66 bffd1780 bfc73f14 8006c37c 8c01825c bfc73f9c 00000002 bffd1700

3ea0: 00000002 bfc72000 bfe8fa38 80038f40 1de968df bfc73edc 00002710 8c01825c

3ec0: 80038f40 00000002 00000002 bffd1780 8076e42c ffffffff bffb4000 00000002

3ee0: 00000000 800909b8 00000000 00000000 00000002 bffb401c bfe8fa00 bfc72000

3f00: bff8ef20 800909dc 8c018f40 bfe8fa00 bfc73fc4 80547770 bfc73f9c bff8c000

3f20: bfc73f44 8006838c 000f4240 00000000 00000001 00000000 1de968df 00000000

3f40: 80038f40 80038f40 800381b4 80038f40 bfc68ae0 bff8be48 8006d9ac 00000001

3f60: bff8be54 00000000 00000000 00000003 bfc73fa4 8006492c ffffffff 00000000

3f80: bff8c000 00000000 bfe8fa00 8008ba48 bfc73f90 bfc73f90 00000031 00000001

3fa0: bfc73fc4 bff8be38 80791a94 803c1958 00000013 00000000 00000000 00000000

3fc0: 00000000 8008b424 80040a64 00000000 80791a94 00000000 00000000 00000000

3fe0: bfc73fe0 bfc73fe0 bff8be38 8008b3a4 80040a64 80040a64 00000000 00000000

[<80064904>] (__wake_up_common+0x28/0x90) from [<80064e74>] (__wake_up+0x3c/0x50)

[<80064e74>] (__wake_up+0x3c/0x50) from [<803c0e94>] (get_res_do_task+0x10c/0x584)

[<803c0e94>] (get_res_do_task+0x10c/0x584) from [<803c1a34>] (ipu_task_thread+0xdc/0xaf0)

[<803c1a34>] (ipu_task_thread+0xdc/0xaf0) from [<8008b424>] (kthread+0x80/0x88)

[<8008b424>] (kthread+0x80/0x88) from [<80040a64>] (kernel_thread_exit+0x0/0x8)

Code: e1a08003 e5b7c008 e59b9004 e157000c (e41c400c)

---[ end trace 831e0a1804e09785 ]---

note: ipu1_task[400] exited with preempt_count 1

mxc_hdmi mxc_hdmi: Read EDID again

mxc_hdmi mxc_hdmi: No modes read from edid

mxc_hdmi mxc_hdmi: create default modelist

pwm_config: pwm freq = 32786, clk_select=2 clock_rate=22000000

pwm_config: pwm freq = 20000, clk_select=2 clock_rate=22000000

imx-ipuv3 imx-ipuv3.0: use special clk parent

ipu_init_sync_panel: disp=0, pixel_clk=27000000 27000000

pwm_config: pwm freq = 32786, clk_select=2 clock_rate=22000000

pwm_config: pwm freq = 20000, clk_select=2 clock_rate=22000000

adv739x_write: reg=0x17 data=0x2 ret=0

adv739x_write: reg=0x0 data=0x1c ret=0

adv739x_write: reg=0x1 data=0x0 ret=0

adv739x_write: reg=0x80 data=0x10 ret=0

adv739x_write: reg=0x82 data=0xcb ret=0

adv739x_write: reg=0x8a data=0x8 ret=0

pwm_config: pwm freq = 32786, clk_select=2 clock_rate=22000000

pwm_config: pwm freq = 20000, clk_select=2 clock_rate=22000000

pwm_config: pwm freq = 32786, clk_select=2 clock_rate=22000000

pwm_config: pwm freq = 20000, clk_select=2 clock_rate=22000000

adv739x_write: reg=0x0 data=0x1f ret=0

imx-ipuv3 imx-ipuv3.0: use special clk parent

ipu_init_sync_panel: disp=0, pixel_clk=27000000 27000000

pwm_config: pwm freq = 32786, clk_select=2 clock_rate=22000000

pwm_config: pwm freq = 20000, clk_select=2 clock_rate=22000000

adv739x_write: reg=0x17 data=0x2 ret=0

adv739x_write: reg=0x0 data=0x1c ret=0

adv739x_write: reg=0x1 data=0x0 ret=0

adv739x_write: reg=0x80 data=0x10 ret=0

adv739x_write: reg=0x82 data=0xcb ret=0

adv739x_write: reg=0x8a data=0x8 ret=0

VPU interrupt received.

I could use some resolution ASAP.

This is kernel 1.1.1 and LTIB 1.1.0

Thanks

Tim

[LeonardoSandova]

Which BSP version are using? which iMX platform?

[timothybean]

Sorry Leonardo,

3.0.35 1.1.0 BSP and IMX6Q

Tim

[LeonardoSandova]

Could you try the same task on 3.0.35 4.0.0 BSP? I remember there was a problem with long playbacks due to a memory leak, but I am not sure if its the same as the one you found.

[timothybean]

I can try, unfortunately, that does not give me much time to actually test the new configuration before I have to ship a unit. Anyways you can research and see if you can find that error so I can patch for the interim?

Tim

[LeonardoSandova]

This is the thread I was taking about (GStreamer crashing on i.MX6 (Boundary Devices Nitrogen6x) ) but reading it seems that the problem was related to big media files, is this your case? Are you using gstreamer? if yes, are you using bin elements, like playbin2?

[timothybean]

Hi Leonardo,

I was able to build the new kernel from Boundary, although, I am back to having issue with decoding video again... HDMI video is all messed up.

Also, trying to build the fs, I get a build fail on glib2...

s_copysignf.c:30:8: error: redefinition of '__copysignf

Any ideas how to resolve this?

Tim

[LeonardoSandova]

Hi Tim,

What is the problem with HDMI playback? No idea regarding the error you are facing on the compilation. You may need to either jump to 4.0.0 or Yocto for latest support. EricNelson can give a much better advice.

Leo

[timothybean]

Problems with IMX6Q dual display and BT656 output

[timothybean]

Yea, that was with the 4.0.0 BSP. I sent Eric an email, waiting for his reply.

You can look at my other post regarding HDMI playback issues. It really seems to be a color space conversion issue brought on by the addition of the BT656 driver.

Tim