--=-=-=
Content-Transfer-Encoding: quoted-printable


On my T60, I just updated to current -current, running GENERIC.MPACPI.
This message is a combination report of what works and what doesn't, a
cross between a request for help and status for others thinking about
a T60.

speedstep shows 1000 and 2167, which are min/max inferred values.  I
can set them with no problems, and time for running make on xplot goes
From=204.9s to 2.3s, so I believe it's working.   It seems to come up at
1000 on battery and 2167 on AC.

sysctl -w machdep.sleep_state=3D1
prints a kernel message that it can't, but nothing bad happens.

sysctl -w machdep.sleep_state=3D3
resulted in

acpi0: entering sleep state 3
pckbd0: suspending...
=3D=3D=3D=3D=3D[acpi_princpu() debug dump]....
5 lines of data
[TOTAL HANG, had to hold power button for ~5 seconds]

Any clues on how to debug this?

I have acpibut0 in dmesg, but powerd doesn't see pushing the button.
How to debug?

acpiacad0 reports the right state on boot (I think), but then often
doesn't get a new state for a long time.  An example was after
unplugging the adapator, then after minutes envstat -r reported a
discharge rate for a battery but still acac connected.  Then, after
perhaps .5 hr to an hour or so it reported disconnected.   After
plugging in it reported a charging battery but still disconnected
adaptor for a long time (at least tens of minutes).

On 'shutdown -p', the file systems are dirty on next boot.  'shutdown
=2Dr' is ok, as is 'shutdown' followed by manual unmounting.  I wonder
if somehow the ACPI poweroff is so fast that the disk queues from the
unmount aren't processed yet.

other hardware status:

The builtin wm(4) sort of works, but PHY programming is clearly wrong.
I have to set media manually to 10baseT and then it will work; even
then after setting to 100baseTX my switch reports it's still at 10.
Now it's finding makphy0 at wm0, but behaves much the same as earlier
current.

After starting X (XFree86, vesa driver, works fine), going back to
text console results in some sort of expanded mode, where only the
upper 1/2 and left 1/2 (more or less) of the screen shows.

Builtin ath(4) and USB work fine.  Bluetooth has worked, and I think
it probably would now - haven't tried recently.  Cardbus works (ath
and Kyocera KPC650 for VZ EVDO).  Sound works with some mixerctl hacks
From=20this list.=20

I had to put the IDE controller into 'legacy' mode in the BIOS, vs
AHCI and then it works fine (UDMA mode 5).

Both processors are found.  I'm running single-core now for more
battery life and because I suspected the bluetooth code a while ago
(perhaps wrongly).

I have not tried the expresscard slot, the (expected) winmodem, the
fingerprint reader, or any TCPA stuff.

    Greg Troxel <gdt@ir.bbn.com>

--=-=-=
Content-Type: application/pgp-signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (NetBSD)

iD8DBQFFLVSv+vesoDJhHiURAm01AJ9RW6tHMh54kpwL/Rg6jb/ZPwBdMgCeK19e
AE/GSP4nU5PBxQlc8sm9AKU=
=zvNw
-----END PGP SIGNATURE-----
--=-=-=--