Old uefi.bin in armv7-lsk-pre-merge-ci boot tar ball not working

Bug #1321905 reported by Naresh Kamboju on 2014-05-21
8
This bug affects 1 person
Affects Status Importance Assigned to Milestone
LAVA Validation Lab
Invalid
Undecided
Unassigned
Linaro CI
New
Undecided
Unassigned
Linaro Stable Kernel (LSK)
Invalid
Undecided
Unassigned
Linaro UEFI
Invalid
Undecided
Unassigned

Bug Description

Old uefi.bin in armv7-lsk-pre-merge-ci boot tar ball not working and due to this vexpress-tc2-benchmark device struck in boot loader.

with below images:
http://snapshots.linaro.org/android//~linaro-android/armv7-lsk-pre-merge-ci/33//boot.tar.bz2
http://snapshots.linaro.org/android//~linaro-android/armv7-lsk-pre-merge-ci/32//boot.tar.bz2

ARM Versatile Express Boot Monitor
Version: V5.1.9
Build Date: Dec 3 2012
Daughterboard Site 1: V2P-CA15_A7 Cortex A7
Daughterboard Site 2: Not Used
Running boot script from flash - BOOTSCRIPT
Fatal Error: Unhandled Exception - Undefined

 <LAVA_DISPATCHER>2014-05-21 02:49:22 PM ERROR: _enter_bootloader failed

 Traceback (most recent call last):
  File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.04-15-g03ffbb0/lava_dispatcher/device/master.py", line 454, in _boot
  self._enter_bootloader(self.proc)
  File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.04-15-g03ffbb0/lava_dispatcher/device/vexpress.py", line 75, in _enter_bootloader
  super(VexpressTarget, self)._enter_bootloader(connection)
  File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.04-15-g03ffbb0/lava_dispatcher/device/target.py", line 422, in _enter_bootloader
  if connection.expect(self.config.interrupt_boot_prompt) != 0:
  File "/srv/lava/.cache/git-cache/exports/lava-dispatcher/2014.04-15-g03ffbb0/lava_dispatcher/utils.py", line 253, in expect
  proc = super(logging_spawn, self).expect(*args, **kw)
  File "/srv/lava/.cache/eggs/pexpect-2.4-py2.7.egg/pexpect.py", line 1316, in expect
  return self.expect_list(compiled_pattern_list, timeout, searchwindowsize)
  File "/srv/lava/.cache/eggs/pexpect-2.4-py2.7.egg/pexpect.py", line 1330, in expect_list
  return self.expect_loop(searcher_re(pattern_list), timeout, searchwindowsize)
  File "/srv/lava/.cache/eggs/pexpect-2.4-py2.7.egg/pexpect.py", line 1414, in expect_loop
  raise TIMEOUT (str(e) + '\n' + str(self))
 TIMEOUT: Timeout exceeded in read_nonblocking().

https://validation.linaro.org/scheduler/job/127630/log_file#L_22_0
https://validation.linaro.org/scheduler/job/127634/log_file#L_20_0

description: updated
affects: lava-dispatcher → lava-lab
Matthew Hart (matthew-hart) wrote :

Please do not add comments or links to into a bug which is clearly a different failure!

Back to the original issue:

We have a "backup" known working uefi.bin which LAVA uses to boot into the master image, this step is working fine.

Before the board is booted into the test image, the uefi.bin from the job specification is written, and this is what seems to be crashing.

I don't believe this is a bug in lava or lava-lab, as it seems to me that the uefi binary in the job is not bootable.

Changed in lava-lab:
status: New → Invalid
description: updated
Ryan Harkin (ryanharkin) wrote :

I agree with Matt, the UEFI binary is crap.

Where did the UEFI binary in that boot.tar.bz2 come from? It's dated 15th Dec 2013 for a start. As are all the UEFI binaries and the uInitrd. Why such old binaries on a new snapshot?

The last Android LSK release had binaries from 20th April 2014.

summary: - vexpress-tc2-benchmark device struck in boot loader
+ Old uefi.bin in armv7-lsk-pre-merge-ci not working
summary: - Old uefi.bin in armv7-lsk-pre-merge-ci not working
+ Old uefi.bin in armv7-lsk-pre-merge-ci boot tar ball not working
vishal (vishalbhoj) wrote :

Premerge CI just picks an old Android Release image and updates the kernel and provides the same to boot. The image that was being used was from 13.12 release. ITS request has been raised to get the image updated to 14.03 now:
https://linaroithelp.zendesk.com/requests/2697

description: updated
Changed in linaro-uefi:
status: New → Invalid
Changed in linaro-stable-kernel:
status: New → Invalid
Chase Qi (chase-qi) wrote :
To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.