Bug 7770

Summary: Errors found on dmesg MM 32 bits firmware for core-sato-sdk image
Product: [Build System, Metadata & Runtime] BSPs Reporter: Costin <costin.c.constantin>
Component: bsps-meta-intelAssignee: California Sullivan <california.l.sullivan>
Status: VERIFIED DUPLICATE QA Contact: Costin <costin.c.constantin>
Severity: normal    
Priority: Medium CC: alexandru.c.georgescu, bogdanx.a.voiculescu, dvhart, warthog9
Version: 1.6.3   
Target Milestone: 1.9   
Hardware: MinnowBoard Max   
OS: x86   
Whiteboard:
OS type for building Yocto: --- Type of Regression: New (Never tested)
Verified: Documentation change: Don't know
Attachments:
Description Flags
for core2-32
none
for corei7
none
logs collected during an oe-seltest run none

Description Costin 2015-05-15 09:01:15 UTC 
Hello,

While performing weekly auto runtime tests on Minnowboard Max, for both 32 and 64 bits UEFI firmware (last current version 0.79 in both cases), found some errors in the auto-test logs (that seem to be similar to those in dmesg, but couldn't find them all there; it might be that they might be fired up only when running auto-tests) and further in dmesg. 

Attached can be found the two auto tests logs. (please ignore the tests themselves and look only at mostly the bottom of the files where system errors appear)

meta-intel: f9738ec19d3b2864c0bd71b9bb3418d16e922a12
poky: 2d09cb28f25080ed337d37d2d4381b846bbe6247 (latest master)
local OS: Ubuntu 15.04

How to reproduce:

1) download core-image-sato-sdk    from upstream, both core2-32 and corei7 and install it on MM

 http://yocto-ab-master.jf.intel.com/pub/releases/yocto-1.8.rc1/machines/

2) set up local env. (by setting the MACHINE) and bitbake rpm psplash package-index
3) solve the issues with directory not found etc when bitbake complains of it and create the necessary .manifest files
4) add the followings in local.conf:
INHERIT += "testimage"
TEST_TARGET = "simpleremote"
TEST_TARGET_IP = "the IP assigned to the BSP"
# Optional
TEST_SERVER_IP = "the IP of the local machine where tests are ran from"
# Optional, but strongly recommended if you want all the tests
TEST_SUITES_append = " auto"
4) add meta-selftest in bblayer.conf
5)build the image as you would normally do:
bitbake core-image-sato-sdk -c testimage | tee autotest.log
6) at the end, go to:
build/tmp/work/intel_core2_32-poky-linux/core-image-sato-sdk/1.0-r0/temp
build/tmp/work/intel_corei7_64-poky-linux/core-image-sato-sdk/1.0-r0/temp
and find the log.do_testimage (that i have attached here)

Costin
Comment 1 Costin 2015-05-15 09:02:10 UTC 
Created attachment 2509 [details]
for core2-32
Comment 2 Costin 2015-05-15 09:02:27 UTC 
Created attachment 2510 [details]
for corei7
Comment 3 Alexandru Georgescu 2015-05-19 15:24:42 UTC 
COstin, can you keep this bug for 32bit only and open another bug for 64bit?

There seems to be diffrent issues in the logs for diffrent hw arch.

Thanks,
Comment 4 Costin 2015-05-19 15:47:13 UTC 
Hello,

Just to summarize for the 32bit firmware MinnowMax, these are the errors that kept my attention during oe-selftest. These errors appeared directly (by themselves) in the oe-selftest log, without taking any special action. I will later on attach syslog, /var/log/messages.

***********************
Central error: [    0.413000] pci 0000:00:1c.0: BAR 14: failed to assign [mem size 0x00200000]
***********************

***********************
Central error: [    0.413024] pci 0000:00:1c.0: BAR 15: failed to assign [mem size 0x00200000 64bit pref]
***********************

***********************
Central error: [    0.412905] pci 0000:00:1c.3: BAR 14: failed to assign [mem size 0x00200000]
***********************

***********************
Central error: [    0.821437] [drm] failed to retrieve link info, disabling eDP
***********************

***********************
Central error: [    0.858378] ahci 0000:00:13.0: controller can't do DEVSLP, turning off
***********************

***********************
Log: /var/log/dmesg
-----------------------
Central error: [    0.412930] pci 0000:00:1c.3: BAR 15: failed to assign [mem size 0x00200000 64bit pref]
***********************

***********************
Central error: [    1.940867] sdhci-acpi 80860F16:00: failed to setup card detect gpio
***********************

***********************
Central error: [    0.413000] pci 0000:00:1c.0: BAR 14: failed to assign [mem size 0x00200000]
***********************

***********************
Central error: [    0.413024] pci 0000:00:1c.0: BAR 15: failed to assign [mem size 0x00200000 64bit pref]
***********************

***********************
Central error: [    0.412905] pci 0000:00:1c.3: BAR 14: failed to assign [mem size 0x00200000]
***********************

***********************
Central error: [    0.821437] [drm] failed to retrieve link info, disabling eDP
***********************

***********************
Central error: [    0.858378] ahci 0000:00:13.0: controller can't do DEVSLP, turning off
***********************

***********************
14 errors found in logs.

Machine information: 
*******************************
Machine name: intel-core2-32
CPU:  Intel(R) Atom(TM) CPU  E3825  @ 1.33GHz
Arch: i686
Physical cores: 2
Logical cores: 2
*******************************

Costin
Comment 5 Costin 2015-05-21 14:18:05 UTC 
Hello,

Attached are logs collected during a oe-selttest run.

Costin
Comment 6 Costin 2015-05-21 14:18:48 UTC 
Created attachment 2522 [details]
logs collected during an oe-seltest run
Comment 7 California Sullivan 2015-07-22 16:31:19 UTC 
These errors appear to be the same as what's the corei7 version, which has been split into multiple bug reports.

Marking as duplicate.

*** This bug has been marked as a duplicate of bug 7793 ***
Comment 8 Costin 2016-01-19 10:26:55 UTC 
Verified as duplicate of bug 7793