ASD halted error with error code(0x800a0004) in Birch Stream

[pengshuang]

I use Cscripts to connect BMC ASD(1.5.2) successfully, but the CPU cannot be halted. The I try to use the Intel XDP tools, it works well.

Here are the logs：

Python 3.10.0 (tags/v3.10.0:b494f59, Oct 4 2021, 19:00:18) 
PythonSv running in interactive mode
win32com not found, using alternative check_running_windows method
Connecting to IPC API....
IPC-CLI: 4.24.21807.100, OpenIPC:5233c3ce87854ac2f44cd05 : 24.2.13416.200
Initializing IPC API....

Info - Detected relevant file modified (or added) after OpenIPC was installed: C:\Intel\OpenIPC\Config\GNR\GNR_ASD_RC-Pins.xml

Info - Detecting remote probes.
Info - OpenIPC configuration 'GNR_ASD_RC-Pins'; run control plug-in(s): 'OpenRC'; probe plug-in(s): 'ASD_JTAG'

Info - ASD plugin connected to: 172.17.0.102:6123 payloadSize: 3000 maxNumBuffers: 20 securityProtocol: None
Info - Chain 0: (Adjusted) JTAG Frequency Divisor: 14
Info - ASD I2C Bus_Select: 0 Bus_Speed: 100000
Info - ASD I2C Bus_Select: 1 Bus_Speed: 100000
Info - ASD I2C Bus_Select: 2 Bus_Speed: 100000
Info - ASD I2C Bus_Select: 3 Bus_Speed: 100000
Info - Hardware GPIO config = 0
Info - JTAG Mode = SOFTWARE
Info - JTAG Chain Select Mode = SINGLE
Info - ASD Probe plugin version: v1.5.8_ASD_BMC_v1.5.2_1.00.0002
Warning - XDP Presence Detected
Info - Added debug port 0 for probe ASD_JTAGProbe

Info - Detected IO_V1_CLTAP A0 on JTAG chain 0 at position 1
Info - Detected IO_V1_CLTAP A0 on JTAG chain 0 at position 0

Info - Survivability Preset was requested, but no supported devices were detected. Preset will be applied when a supported device is detected.
TargetEvent: Power : On -- 15:41:45.442485
TargetEvent: PowerDomain : PCH : On -- 15:41:45.443485
TargetEvent: PowerDomain : CPU : On -- 15:41:45.443485
Target System: GNR - GraniteRapids
Running Graniterapids
Using ASD_JTAGProbe as the OpenIPC connection.
Creating SOC CHA and Core logical view...
Current Working Dir: C:\Users\46143\PythonSv

*******************************************************************************
Welcome to CScripts for Server & Micro-server platforms!
These scripts are provided as-is as a courtesy to assist with platform
debug. They are NOT intended for manufacturing or production line validation,
and Intel makes no guarantee, implied or otherwise, that these scripts will
function in a production environment. They are not fully validated by Intel
nor guaranteed to execute as described. These scripts rely on (in some cases)
unspecified features within our components which may or may not be functional
from stepping to stepping, or from product to product. They are provided at
no cost. Customers are free to modify these scripts as needed for their own
platform debugging purposes. Defects should be reported to your Intel account
representative.

All log files will be placed in C:\Users\46143\PythonSv
For help on functions or objects when in the shell type help(NAME)

*******************************************************************************

Cscripts version: BHS.2422.6001
IPC-CLI Version 4.24.21807.100
Probe Information: v1.5.8_ASD_BMC_v1.5.2_1.00.0002
Start Platform Stuff with access type ipc
WARNING!!! ltssm_gnr should not be imported directly. Import pysvtools.pciedebug.ltssm instead
Socket 0: detected Server-SP, 80 cores, A2, PCI Segment 0, Bus List = [0x00, 0x01, 0x03, 0x05, 0x06, 0x08, 0x0a, 0x0b, 0x0d, 0x0f, 0x10, 0x12, 0x14, 0x29, 0x3a, 0x4b, 0x5c, 0x6d, 0x7e, 0x7f]
Socket 1: detected Server-SP, 80 cores, A2, PCI Segment 0, Bus List = [0x80, 0x81, 0x83, 0x85, 0x86, 0x88, 0x8a, 0x8b, 0x8d, 0x8f, 0x90, 0x92, 0x94, 0xa6, 0xb7, 0xc8, 0xd9, 0xea, 0xfe, 0xff]
Python 3.10.0 (tags/v3.10.0:b494f59, Oct 4 2021, 19:00:18) 
Type 'copyright', 'credits' or 'license' for more information
IPython 8.26.0 -- An enhanced Interactive Python. Type '?' for help.

In [1]: halt()
---------------------------------------------------------------------------
IPC_Error Traceback (most recent call last)
Cell In[1], line 1
----> 1 halt()

File cscripts\dependencies\ipccli\ipc_env\ipc_baseaccess.py:1291, in halt(self)

File cscripts\dependencies\py2ipc\IPC2PyUtils.py:105, in newfunc(self, *args, **kwargs)

File cscripts\dependencies\py2ipc\Service_OperationReceipt.py:168, in Flush(self, token)

File cscripts\dependencies\py2ipc\IPCTypes.py:83, in IPC_ErrorCheck(result)

IPC_Error: RunControl_Timeout == 0x800a0004
. Timeout occurred while waiting for threads to halt with 0 retries

[Subhashish]

Hello pengshuang,

Thank you for your post. We are reviewing these details. While we are working on this, we would like you to share the further information like processor model, system board details like model, part numbers. These would be really helpful for us to analyze this further.

Thank you for choosing Intel.

Regards,

Subhashish_Intel.

[pengshuang]

Hi Subhashish,

Thank you for your reply. The processor model is GENUINE INTEL(R) XEON(R) and the system board model is T4DM based on Intel CRB BeechnutCity.

[Fikri_Intel]

Hi pengshuang,

Thank you for your response.

Upon checking, the system board model is T4DM designed by ttyinfo based on Intel CRB BeechnutCity, We kindly advise you to open an IPS case for further support.

Please open an IPS case at premiersupport.intel.com and after registration, you will be able to open a case on the Intel® Premier Support (IPS) platform and your request will be handled by one of our engineers as soon as possible. 

If there is no further concern, we will proceed to close this thread, and no longer will be monitored.

Regards,

Fikri

[Subhashish]

Hi pengshuang,

We hope you were able to connect with IPS team from the  premiersupport.intel.com after registration. As there is no action required from our end, we are proceeding to close this ticket.

Thank you.

Regards,

Subhashish_Intel.

[pengshuang]

OK, Thank you.

[AaronWu]

Hi,

We also got the same error code in CRB board with CScripts in 686308 & use the ISD w_sys_dbg_nda_p_2024.2442.0.88_offline.exe.

Does this problem have solution?

In [1]: itp.halt()
---------------------------------------------------------------------------
IPC_Error Traceback (most recent call last)
Cell In[1], line 1
----> 1 itp.halt()

File cscripts\dependencies\ipccli\ipc_env\ipc_baseaccess.py:1291, in halt(self)

File cscripts\dependencies\py2ipc\IPC2PyUtils.py:105, in newfunc(self, *args, **kwargs)

File cscripts\dependencies\py2ipc\Service_OperationReceipt.py:168, in Flush(self, token)

File cscripts\dependencies\py2ipc\IPCTypes.py:83, in IPC_ErrorCheck(result)

IPC_Error: RunControl_Timeout == 0x800a0004
. Timeout occurred while waiting for threads to halt with 0 retries. Error occurred in enDebug: enDebug not supported.

[enarvaez]

Hi,

Am using cscripts BHS.2334.6000 + XDP vs AvenueCity and have the same RunControl_Timeout error.   (From what I hear from Intel is that error is common).   Am in progress working with Intel.   

First we used RWEverything installed on AvenueCity to verify MSR 0xC80.0 last bit=0), which states that CPU_Debugging is disabled.   They also wanted to verify CPU silkcreen (Q416, J4020020).  

Then read "bin file" from CPU1 port with dediprog, as described in 636255_AvenueCity_TOPS_rev1-2.pdf.  Opened in FITm, though CPU_Debugging currently greyed out and not configurable.

Now the ask is to provide "bin file" from the "working AvenueCity".  

Lots of hoops to jump through so far....   Will advise.   Appreciate any insight as well. 

Intel 

Also asked to update cscripts to 2431.6000.   I installed on the same machine, but am getting a "magic number" error.   Seen this issue before, previously fixed by ensuring OpenIPC/cscript/python version match and to fixed with dfmanager.   Though currently is failing to create virtual environment with a different error.   

[enarvaez]

Resolved after working with Intel SDP.   Not an issue with cscript/isd.  IPC_Error: RunControl_Timeout == 0x800a0004 due to BIOS strapping defaulted has inability to itp.halt().  Solution was to load BIOS in FITm tool, toggle straps (below shows what worked), save/flash new BIOS.

CPU Debugging - Disabled
CONSENT - Enabled
DAM – Enabled
TXT - No

Magic Number issue truly issue with python mismatch, configurable with dfmanager.

[Alick]

Hi enarvaez,

Can you share link where I can find the instruction to do the following, please?

Solution was to load BIOS in FITm tool, toggle straps (below shows what worked), save/flash new BIOS.

CPU Debugging - Disabled
CONSENT - Enabled
DAM – Enabled
TXT - No

[Grady1]

Hi,

when setup the IO Margin tool for test,can't sussesceful to test,

I go to check it by ipccli,can find the Jtag & get the system info ,

but can't hold the cpu to control its (already attch log for review  )

------------------------------------------- 

CPUs x2 Granite Rapids Server SP ES2-Q4GN 

IO Margin tool version :1.3.37 

system debugginf tool :2440

check the BIOS setup followed  it enable /disable 

grady ,

thanks for u