SNVS RTC fault on imx6q

[lqw39]

Hi,

After changing another batch of imx6q chip, we found android system cannot boot done, it always hang at:

......

I/Zygote  ( 2151): ...preloaded 379 resources in 1771ms.

I/Zygote  ( 2151): ...preloaded 31 resources in 9ms.

D/dalvikvm( 2151): GC_EXPLICIT freed 29K, 1% free 6467K/6531K, paused 2ms+2ms

D/dalvikvm( 2151): GC_EXPLICIT freed 10K, 2% free 6457K/6531K, paused 1ms+1ms

D/dalvikvm( 2151): GC_EXPLICIT freed <1K, 2% free 6457K/6531K, paused 2ms+2ms

I/dalvikvm( 2151): System server process 2225 has been created

I/Zygote  ( 2151): Accepting command socket connections

D/SystemClock( 2225): Setting time of day to sec=86400

W/SystemServer( 2225): System clock is before 1970; setting to 1970.

the batch number with this issue is:

MCIMX6Q6AVT10AC SBCF1315 KOREA YOCFSBU.

the following chip has no problem:

MCIMX6Q6AVT10AC SBAP1250 KOREA XXAPSBV.

[Update] We used RTC stub instead of SNVS for test, and found system can boot up, but run very slowly. It is almost certainly SNVS is going wrong.

Anybody have any ideas?

Thanks.

[LinWang]

Dear Mr. Liu,

Based on above info which you had provided.

We can't get some conclusion.

Because the different chip batch will not bring any dfiierent.

But I have below suggestions fro you.

1. Please check your SW version, is it latest one and added all patched?

2. Please check power supply status and sequence.(VDDSNVS_IN should power up first.)

Hpoe above info can help you!

[aleksandrmuravi]

Dear Mr. Liu,

Based on above info which you had provided.

We can't get some conclusion.

Because the different chip batch will not bring any dfiierent.

But I have below suggestions fro you.

1. Please check your SW version, is it latest one and added all patched?

2. Please check power supply status and sequence.(VDDSNVS_IN should power up first.)

Hpoe above info can help you!

Sory if my question is stupid, but what do you mean when say "SW version"?

Regards.

[lqw39]

Hi Lin,

We have applied all patches listed on website, and use default power sequence of PFUZE100 (same with reference design, official BSP) .

BTW: system does not crash, just halt, and "android" animation still continues

As I know, some batches of i.MX51 chip do not support security RTC module (or shielded?), so customers have to use RTC in PMIC (MC13892); unfortunately, we encountered this situation.

[LinWang]

Dear Mr. Liu,

Can you show me your schematic?

BTW, I can't understand your below descriptions.

"[Update] We used RTC stub instead of SNVS for test" and "system does not crash, just halt, and "android" animation still continues"

All i.Mx6 chip with SRTC module, you can find related info in Datasheet.

Thanks!

[lqw39]

We loaded a simulated RTC driver instead of SNVS RTC driver in kernel, and found system can run normally.

BTW: i.MX515 datasheet says SRTC is supported, but some or all of them shipped into China do not support SRTC, we have confirmed this issue with FAE of WT Microelectronics Co. Ltd, which is one of the Freescale Distributor.

[LinWang]

Dear Mr. Liu,

Do you mean no issue after using your simulated RTC driver now?

i.Mx6 maybe different i.Mx515, Please confirm i.Mx6 status again.

Thanks!

[lqw39]

Yes, android system works well after using simulated RTC driver; but media scanning becomes slow.

We'll check other i.MX6 status.

Thanks.

[YixingKong]

Qinggwei, please click Correct Answer/Helpful Answer, if your question has been answered.

Thanks,

Yixing

[lqw39]

1. We have confirmed that clock source is right, and the simulated RTC does not change clock settings.

2. We cannot use simulated RTC in product since time cannot be kept after system power down; we MUST make SNVS RTC working.

Some batches of imx6q chip do not support SRTC?

Shall we add new RTC chip into our hardware design?

[LinWang]

As I know, RTC function is availabe in all i.Mx6 batches.

Please list your BSP version and pach list which you had added.

I will ask our SW engineer to help checking.

Thanks!

[lqw39]

We have tried following BSP version:

1. R13.4

2. R13.41 + IMX6_R13.4.1.04_ANDROID_PATCH

3. JB-422_100

and every version have this issue on specific chip.

[LinWang]

How many chip/board have this issue?

We need conifrm this isn't a single chip/board issue.

Please share your schematic with me.

I like check it at same time.

[lqw39]

All of the 10 pcs (with chip SN: MCIMX6Q6AVT10AC SBCF1315 KOREA YOCFSBU) have this issue.

And other boards (with chip SN: MCIMX6Q6AVT10AC SBAP1250 KOREA XXAPSBV) have no issue.

Two kinds of board have no difference except i.mx6q chip.

We used the same design with SDP board.

Thanks.

[LinWang]

Please try latest BSP.

let me know the result.

Thanks!

[lqw39]

As I said above, we have tried the latest version we can get from website.

If you have newer version, please give me the link.

Thanks.

[LinWang]

Sorry, I miss your latest important info!

Suggestions:

1. Find a FSL domo board and mount issue chip "1315" on it an test.

or

2. Find local FSL support for help.

Would you please show your company's name and office location to us?

Thanks!

[RobinGong]

It seems your chips were not programmed with right fuse bit. Can you type "imxotp read 4" and let me know the value in u-boot ?

[lqw39]

The result of "imxotp read 4" is:

MX6Q SABRESD U-Boot > imxotp read 4

Reading fuse at index: 0x4

Fuse at (index: 0x4) value: 0x0

[RobinGong · Accepted solution]

Yes, the bit1 (SEC_CONFIG[0] )of this register is 0 which means SNVS function is disabled , so that rtc module can’t work as expect. You can blow the right fuse bit on one chip to test if it can fix the issue. Please type “imxotp blow –force 0x2A0302” in u-boot.

[aleksandrmuravi]

Dear RobinGong.

I try blow SEC_CONFIG[0] to 1, but get the error

MX6Q SABRESD U-Boot > imxotp blow --force 0x2A0302

imxotp - One-Time Programable sub-system

Usage:

imxotp imxotp read <index>

- read fuse at 'index'

imxotp blow [--force] <index> <value>

- blow fuse at 'index' with hex value 'value'

Tips:

'addr' to 'index':

convert 'index' from 'address'

index = (addr - otp_base) / 0x10

eg, addr is 0x021bc410, otp_base is 0x021bc400, the index = 1

Can you help me with this issue?