#12727 NORM Not Tri: OFW doesn't follow x86 linux boot protocol
Zarro Boogs per Child
bugtracker at laptop.org
Fri Jul 12 22:14:07 EDT 2013
#12727: OFW doesn't follow x86 linux boot protocol
-------------------------------------------+--------------------------------
Reporter: dsd | Owner: Quozl
Type: defect | Status: new
Priority: normal | Milestone: Not Triaged
Component: ofw - open firmware | Version: not specified
Resolution: | Keywords:
Next_action: communicate | Verified: 0
Deployment_affected: | Blockedby:
Blocking: |
-------------------------------------------+--------------------------------
Changes (by Quozl):
* next_action: never set => communicate
Comment:
Please test with crank:~quozl/q3c16ja.rom which:
* clears the sentinel byte at 0x1ef, and
* reports whether the sentinel clearing is needed or not, based on the
kernel header version number being greater than 0x020b, per 09c205a,
{{{
Trying last:\boot\bootfw1.zip
FW found - current FW is up-to-date
Trying last:\boot\runos1.zip
OS found - kernel header version number check, sentinel wiping is
superfluous
Trying last:\boot\runrd1.zip
RD found - [ 0.000000] Initializing cgroup subsys cpu
}}}
q3c16ja is q3c16 plus [http://dev.laptop.org/~quozl/z/1UxpF8.txt this
patch].
code16-size reads the byte at 0x1f1 in the loaded kernel image, adds one,
and multiplies by 512. This is setup_sects, the size of the setup in
sectors.
I tried to build a kernel here, but the olpc-3.10 branch asked me many
config questions that I had no answers for, after copying xo_1.5_defconfig
to .config. If q3c16ja doesn't fix it, could you provide a test kernel?
What is the symptom of the failure? Presumably missing /proc/device-tree
and related symptoms.
--
Ticket URL: <http://dev.laptop.org/ticket/12727#comment:1>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system
More information about the Bugs
mailing list