From: Abel Abraham Camarillo Ojeda <acamari@verlet.org>
Subject: Re: Thinkpad t410 does not reboot properly
To: Walter Alejandro Iglesias <wai@roquesor.com>
Cc: tech@openbsd.org
Date: Sun, 9 Jun 2024 05:42:32 -0600

On Sun, Jun 9, 2024 at 5:15 AM Abel Abraham Camarillo Ojeda <
acamari@verlet.org> wrote:

> On Sun, Jun 9, 2024 at 5:03 AM Walter Alejandro Iglesias <wai@roquesor.com>
> wrote:
>
>> On Sun, Jun 09, 2024 at 12:39:53PM +0200, Walter Alejandro Iglesias wrote:
>> > > Well, I guess it depends in the amount of changes that occurred
>> during that
>> > > day,
>> > > if you pinpointed a change already I guess its not needed
>> >
>> > If its about the kernel, searching in cvsweb by date the nearest
>> > modification I found is this:
>> >
>> >
>> https://cvsweb.openbsd.org/src/sys/arch/amd64/amd64/pmap.c?rev=1.168&content-type=text/x-cvsweb-markup&sortby=date
>> >
>> > Please note that I am no expert, I have no idea what the problem may be
>> > related to.
>> >
>>
>> Thank you Abel!
>>
>
> Hi Walter,
>
> Checking the time frame I see the following changes (mangled by gmail),
> I think any of these could be the responsible one. The times I have
> bisected I did it
> using a binary search algorithm:
>
> $ TZ=UTC cvs up -D '2024/05/02 00:00' -Pd
>
> then
> $ TZ=UTC cvs up -D '2024/05/03 00:00' -Pd
>
> then
> TZ=UTC cvs up -D '2024/05/02 12:00' -Pd
>
> if it works OK then:
> TZ=UTC cvs up -D '2024/05/02 18:00' -Pd
>
> if it fails then:
> TZ=UTC cvs up -D '2024/05/02 15:00' -Pd
>
> and so on...
>

Reading the commit messages my bet would be on this:

$ TZ=Z cvs log -N -d'20240602<20240603' dev/acpi/acpi.c  dev/acpi/amltypes.h
RCS file: /cvs/src/sys/dev/acpi/acpi.c,v
Working file: dev/acpi/acpi.c
head: 1.430
branch:
locks: strict
access list:
keyword substitution: kv
total revisions: 430;   selected revisions: 1
description:
----------------------------
revision 1.430
date: 2024/06/02 11:08:41;  author: kettenis;  state: Exp;  lines: +9 -1;
 commitid: McNWbgdeepOyZIw5;
Power down PCI devices in S0 as well.  Needed for suspend-to-idle.

ok mglocker@
=============================================================================

RCS file: /cvs/src/sys/dev/acpi/amltypes.h,v
Working file: dev/acpi/amltypes.h
head: 1.50
branch:
locks: strict
access list:
keyword substitution: kv
total revisions: 50;    selected revisions: 1
description:
----------------------------
revision 1.50
date: 2024/06/02 11:08:41;  author: kettenis;  state: Exp;  lines: +2 -1;
 commitid: McNWbgdeepOyZIw5;
Power down PCI devices in S0 as well.  Needed for suspend-to-idle.

ok mglocker@
=============================================================================

So:

$ TZ=UTC cvs up -D '2024/05/02 11:07' -Pd

Will get you just before that commit and

$ TZ=UTC cvs up -D '2024/05/02 11:09' -Pd

Just after.

Maybe CC kettenis, I guess he will ask (again) for full sendbug -P output
(including ACPI data)

regards~