E3815 custom board truble

[idata]

I developed board on E3815. But I can't boot it.. please review my trouble:

1. I have minnow turbo and successfully made coreboot custom bios for it (with DCT, fsl and other). I add print POST code to debug uart - and see it.

2. when I write my bios (and other ) in my board I see that E3815 read it from flash after reset (CPU is S0 state) and it doesn't do anything. I don't see anything on debug uart..

I have XDP3 board and I can connect it to my board.. But I can't understand what I can do whit it...

Please help me..

[CarlosAM_INTEL]

Hello, zhgulev:

Thank you for contacting Intel Embedded Community.

In order to better understand this situation, we need to address the following questions:

Could you please tell us with the assistance of your BIOS vendor if affected BIOS implementation fulfills the suggestions stated in the Intel(R) Atom(TM) processor E3800-M/D SoC: BIOS Writer's Guides? These documents can be accessed if you have a CDNA between your company and Intel as well as a Privileged EDC account because they are classified as Intel confidential. To request an upgrade from your Basic EDC account to a Privileged account, go to http://www.intel.com/content/www/us/en/embedded/embedded-design-centersupport.html http://www.intel.com/content/www/us/en/embedded/embedded-design-centersupport.html and click on "Manage Your Intel Profile" found in the "Manage Your Account" section of the page. They can be found at:

https://edc.intel.com/Link.aspx?id=7009 Intel(R) Atom(TM) processor E3800-M/D SoC: BIOS Writer's Guide (Volume 1 of 2) document # 514147

https://edc.intel.com/Link.aspx?id=7010 Intel(R) Atom(TM) processor E3800-M/D SoC: BIOS Writer's Guide (Volume 2 of 2) document # 514148

https://edc.intel.com/Link.aspx?id=7011 Intel(R) Atom(TM) SoC E3800-I: BIOS Writer's Guide Addendum document # 526998

Could you please confirm if the affected design has been implemented based on the https://edc.intel.com/Link.aspx?id=7001 Intel(R) Atom(TM) Processor E3800 Product Family Platform: Design Guide (PDG) document # 512379 recommendations? This PDG requires an EDC privileged account to be accessed as well.

Please give us all the information that may answer the previous questions.

Thanks for your cooperation to solve this inconvenience.

Best regards,

Carlos_A .

[idata]

Now I try to do this - https://software.intel.com/en-us/articles/how-to-check-and-read-the-postcode-in-early-boot-up-stage-by-intel-system-debuggerr How to check and read the postcode in early boot-up stage by Intel System Debugger(R) | Intel® Software

I've got some results:

1. when i turn of TXE region in bios image both my device connect to XDP but tell "debug is off"

2. when I turn on TXE region in bios:

- minnow board connect to debugger and successfully work ( i can set up breakpoint on 0x80 port - and stop on it)

- custom board connect to debugger - but disassembler windows is clear. WHY?

and other things -why I see 2core in studio? and I don't see E3815 in device list (when I select board type).

Now I try to get full access

[CarlosAM_INTEL]

Hello, zhgulev:

Thanks for your update.

Please let us know as soon as you have access to the recommended documents and the answer to our previous questions.

We really appreciate your cooperation.

Best regards,

Carlos_A.

[BTest]

zhgulev,

The boot process of this chip does a lot of things before it releases the x86 core which leads to a lot of confusion. As you have seen, there is a lot of activity to the boot flash but if you look at what is read it is just the descriptor table and the TXE. One thing that seems to have gotten multiple people, including the minnowboard designers, is you must have pull-ups on the SVID_ALERT, SVID_CLK and SVID_DATA lines otherwise the hardware never gets acknowledgement that VCORE is OK and holds the x86 in reset. So while you think the chip is running the part you want is not. JTAG works even if x86 is not. And as you have found out, you must have the TXE loaded to do debugging to the x86 core.

Intel will do a schematic review for you for free that finds things like this. If you haven't done so already I strongly suggest it.

Other areas to look at revolve around the RTC. Even if you are not using it you still have to supply a clock to it.

Let me know if I can be of further assistance.

Brett

[NZhgu1]

Brett,

Thanks a lot for your advice. We will check it faster as can. At the first blush - all ok.. but we will check it very carefully.

[BTest]

If you don't have a copy of 537072_537072_BayTrail_I_Schematic_Checklist then make sure you get one from Intel. It covers a lot of items like SVID.

I went though this 18 months ago and couldn't have done it without the support of people on this forum. The main problems we have run into are getting all the voltage regulators working correctly. Things you don't normally think of, like the DC resistance of an inductor, can be a show stopper.

[NZhgu1]

Brett could you send it to my email mailto:Nikolay.Zhgulev@njm-project.ru Nikolay.Zhgulev@njm-project.ru

Can I ask you some question in skype?

[BTest]

Its a "Intel Confidential" document so you have to get it from Intel. Sorry.

I really can't do skype as I am at work right now, but you can contact me through my personal email, mailto:btest05@gmail.com btest05@gmail.com if you want to take this offline.

Brett

[NZhgu1]

Brett we check SVID lines and found one assembled error.. After corrected it we have had strange result - our XDP3 cant connect to our boadr:

E-L-C-CONERR: target connection failure. E-2201: Unable to connect to target: Target IDCODE does not match the expected value!

Please verify the selected scanchain matches the actual target or override this check.

Expected: 0x0A69A013 Received:

All bits in the expected mask must be set, extra stepping bits are ignored.

Why it happened?

we continue chek this interface and I hope today I will have access to confidential documentation...

[NZhgu1]

Brett we checked all thing from (537072_537072_BayTrail_I_Schematic_Checklist) - all ok..

I start to read documentation about writing bios.

[gverm2]

Hello Bret, We are also same problem of processor booting...CPU is not coming out of reset... I have crosscheced the schematic.....

I am running minnowboard bootloader( for e3826) on my board based on E3815... Is the cause of problem?

Please help me out...

[CarlosAM_INTEL]

Hello, gaurav7931:

Thank you for contacting Intel Embedded Community.

In case that your question is related to the Carlos_A .

[NZhgu1]

Now I have next situation... Board start .. but memory doesn't work.. I don't see any activity on DDR interface.. I have 8chip in 8bit mode..

is there any idiea?

I use fsp V4_debug

============= PEIM FSP (VLYVIEW0 0x00000304) =============

Register PPI Notify: DCD0BE23-9586-40F4-B643-06522CED4EDE

Install PPI: 8C8CE578-8A3D-4F1C-9935-896185C32DD3

Install PPI: 5473C07A-3DCB-4DCA-BD6F-1E9689E7349A

The 0th FV start address is 0x000FFFE0400, size is 0x00017C00, handle is 0x0

Register PPI Notify: 49EDB1C1-BF21-4761-BB12-EB0031AABB39

Register PPI Notify: EA7CA24B-DED5-4DAD-A389-BF827E8F9B38

Install PPI: B9E0ABFE-5979-4914-977F-6DEE78C278A6

Install PPI: DBE23AA9-A345-4B97-85B6-B226F1617389

юInstall PPI: 06E81C58-4AD7-44BC-8390-F10265F72480

Install PPI: 01F34D25-4DE2-23AD-3FF3-36353FF323F1

Install PPI: 49EDB1C1-BF21-4761-BB12-EB0031AABB39

Notify: PPI Guid: 49EDB1C1-BF21-4761-BB12-EB0031AABB39, Peim notify entry point: FFFE0FD4

The 1th FV start address is 0x000FFFB0000, size is 0x0002F400, handle is 0xFFFB0000

ЬInstall PPI: A55D6970-1306-440C-8C72-8F51FAFB2926

ЬPcdMrcInitTsegSize = 8

PcdMrcInitMmioSize = 800

PcdMrcInitSPDAddr1 = A0

PcdMrcInitSPDAddr2 = A2

Setting BootMode to 0

Install PPI: 1F4C6F90-B06B-48D8-A201-BAE5F1CD7D56

Register PPI Notify: F894643D-C449-42D1-8EA8-85BDD8C65BDE

About to call MrcInit();

BayleyBay Platform Type

RID = 0x11.

Reg_EFF_DualCH_EN = 0x40030040.

CurrentMrcData.BootMode = 4

Configuring Memory Start...

[NZhgu1]

according to the document https://deakons.fedorapeople.org/tn4107_power_up_initialization_and_reset.pdf https://deakons.fedorapeople.org/tn4107_power_up_initialization_and_reset.pdf DRAM0_CKP signal must be set to an active state. But in our design, I can only see the transition DRAM0_DRAMRST# to 1, but I do not see that DRAM0_CKP changes its state. On Minnow DRAM0_CKP becomes active.

What can affect the DRAM0_CKP ?

[NZhgu1]

After I installed 2014SP1 and set threadcount=1. After this I could connect to my board.

Now I try configure EDK2 source for compile it fo my board..

Are the Anybody now how I can set E3815 profile in EDK (cash size and count cpu core)..

[JDood]

Hi njm & zhgulev,

My organisation, Ircona (www.ircona.com), area design consultancy firm who have been developing hardware and BIOS solutions for our customer for over 15 years and provide design support for companies who run into difficulties in the development or bring-up of their x86 platforms. As Brett mentioned above, the bring-up process of an x86 processor is very complicated and can require broad hardware and software knowledge to bring it to a successful conclusion.

If you would like to discuss how Ircona can assist you, please email me at mailto:john.doody@ircona.com john.doody@ircona.com

Thanks

John Doody

CEO

Ircona

[Adolfo_S_Intel]

Hello zhgulev

The best way to set up the profile for E3800, is to use the Intel FSP kit for the E3800.

The FPS kit can be downloaded here:https://github.com/latelee/coreboot/tree/master/intel/fsp/baytrail coreboot/intel/fsp/baytrail at master · latelee/coreboot · GitHub

The FSP can be configured to match the design of your board using the Binary Configuration Tool:https://github.com/IntelFsp/BCT GitHub - IntelFsp/BCT: Binary Configuration Tool for Intel(R) FSP

FSP can take care of the SoC chipset initialization, here you can find documentation about how to use FSP with EDKII: https://firmware.intel.com/learn/fsp/about-intel-fsp Intel® FSP - About | Intel® Architecture Firmware Resource Center

Best regards,

Adolfo Sanchez.

[NZhgu1]

Unfortunately standard way doesn't work.. because we have original memory configuration .

And we have trouble with DDR3 initialitation..