Re: i915 hangs after 201612251500Z changes

[Cherry G. Mathew <cherry%zyx.in@localhost>]

>>>>> "Jun" == Jun Ebihara <<jun%soum.co.jp@localhost>,> writes:

    Jun> From: Martin Husemann <martin%duskware.de@localhost>
    Jun> Subject: Re: i915 hangs after 201612251500Z changes
    Jun> Date: Fri, 6 Jan 2017 09:44:46 +0100

    >> On Fri, Jan 06, 2017 at 05:06:03PM +0900, Jun Ebihara wrote:
    >>> > 2016-12-26 23:00 boot fine
    -> UTC 2016-12-26 14:00 boot fine
    >>> > 2016-12-27 00:00 failed
    -> UTC 2016-12-26 15:00 boot failed

    >> The "cvs timestamp" theory is likely, btw., there was an issue on the server
    >> that christos fixed yesterday.
    >> Did you clean the kernel build directory between iterations?

    Jun> thanx for advices,

    Jun> src/sys/arch/i386/i386: machdep.c 1.774->1.775 change affects my i915.

    Jun> /*      $NetBSD: machdep.c,v 1.774 2016/12/23 09:24:57 maya Exp $       */
    Jun> works well.

    Jun> dmesg:
    Jun> https://github.com/ebijun/NetBSD/blob/master/dmesg/i386/Acer_AspireONE_ZG8

    Jun> update machdep.c,v 1.775

    Jun> /*      $NetBSD: machdep.c,v 1.775 2016/12/26 15:47:48 cherry Exp $     */

    Jun> http://mail-index.netbsd.org/source-changes/2016/12/26/msg080203.html
    Jun> http://cvsweb.netbsd.org/bsdweb.cgi/src/sys/arch/i386/i386/machdep.c.diff?r1=1.774&r2=1.775&f=h

    Jun> boot failed,screen turns black and hangs.


Hello Jun,

coypu@ pointed me to this - it's a peculiar one - to see why, do the
following:

cvs rdiff -u -r1.772 -r1.775 src/sys/arch/i386/i386/machdep.c

You'll see that it's almost line for line equivalent to the previous
code. In fact, -r1.774 was incorrect, because it assumed that memory
blocks were next to each other.

I'm trying to figure out if any of the assumptions that I've made are
drastically different. Does your laptop have a peculiar RAM layout ? 

Do you have an e820 map that can be shared ?

Many Thanks,
-- 
~cherry