On 8/24/2013 11:19 PM, Jean Delvare wrote:
Hi Guenter,
On Fri, 23 Aug 2013 04:28:01 -0700, Guenter Roeck wrote:
On 08/23/2013 02:53 AM, Jean Delvare wrote:
On Fri, 23 Aug 2013 10:01:14 +0200, Jean Delvare wrote:
On Thu, 22 Aug 2013 09:13:36 -0700, Guenter Roeck wrote:
On Thu, Aug 22, 2013 at 06:09:01PM +0200, Toralf Förster wrote:
Why is that module [coretemp] not auto loaded ? Hhm ?
It never was. Maybe your relase update replaced /etc/modules.
Actually it was, since kernel v3.4 and commit 9b38096fd. And it works
Me surprised. Well, we live and learn. Guess that was due to introducing
MODULE_DEVICE_TABLE ?
Yes, exactly.
for me with kernel v3.10, but no longer with kernel v3.11-rc6. In other
words I see the same regression Toralf is seeing. I have no explanation
for it at this point, but I have a few candidate commits for reversal.
I'll post again when I know more.
Reverting my 4 candidates did not help. We're up for a bisection I guess,
when I can find the time for it (i.e. not right now.)
I don't see a difference in the driver [coretemp] since 3.10 which would
explain the problem. If it is a generic problem, it would be surprising that
no one else as seen it with other modules.
Might be because CPU driver autoloading is still fairly recent so most
distributions don't rely on it yet. As a matter of fact I had to take
coretemp out of /etc/sysconfig/lm_sensors in order to reproduce the bug.
Bisection pointed to:
commit ac212b6980d8d5eda705864fc5a8ecddc6d6eacc
Author: Rafael J. Wysocki <rafael.j.wysocki@xxxxxxxxx>
Date: Fri May 3 00:26:22 2013 +0200
ACPI / processor: Use common hotplug infrastructure
Rafael, any hint as to why the auto-loading of driver coretemp would be
broken by the aforementioned commit?
Essentially, that is a udev issue.
There is a line like this in the udev rules:
DRIVER!="?*", ENV{MODALIAS}=="?*", IMPORT{builtin}="kmod load
$env{MODALIAS}"
which causes modalias to be matched only for devices that have no
drivers (usually, that line is located in the 80-drivers.rules file
under /lib/udev/rules.d/ or /usr/lib/udev/rules.d/). After the commit
above the CPU devices have a driver and the modalias matching is not
used for them, unfortunately.
The easiest way to fix this is to remove the driver check from that line
so that it looks like this:
ENV{MODALIAS}=="?*", IMPORT{builtin}="kmod load $env{MODALIAS}"
That change has been made in the upstream udev already as far as I can
say, so it should propagate through updates.
Thanks,
Rafael
---------------------------------------------------------------------
Intel Technology Poland sp. z o.o.
z siedziba w Gdansku
ul. Slowackiego 173
80-298 Gdansk
Sad Rejonowy Gdansk Polnoc w Gdansku,
VII Wydzial Gospodarczy Krajowego Rejestru Sadowego,
numer KRS 101882
NIP 957-07-52-316
Kapital zakladowy 200.000 zl
This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.
_______________________________________________
lm-sensors mailing list
lm-sensors@xxxxxxxxxxxxxx
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
