linux-kernel - Re: [lm-sensors] it87 sensors need an ACPI driver (2.6.31)

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <68676e00910060800h4cb84dcl9f610f6b23a88e5c@mail.gmail.com>
Date:	Tue, 6 Oct 2009 17:00:11 +0200
From:	Luca Tettamanti <kronos.it@...il.com>
To:	Michael Tokarev <mjt@....msk.ru>
Cc:	Brad Campbell <brad@...p.net.au>, lm-sensors@...sensors.org,
	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [lm-sensors] it87 sensors need an ACPI driver (2.6.31)

On Mon, Oct 5, 2009 at 6:42 PM, Michael Tokarev <mjt@....msk.ru> wrote:
> Luca Tettamanti wrote:
> []
>>>
>>> Vanilla Kernel 2.6.31.1 fails to load it87 on my machine with the same
>>> messages as posted above.
>
> []
>>>
>>> In addition, asus_atk0110 loads but leaves no trace in dmesg when loaded.
>>
>> That's good ;-) The driver is quiet when everything is ok. You should
>> see the corresponding entries under /sys/class/hwmon/hwmonN; upgrade
>> lm-sensor package to version 3.1.1, older versions do not support ACPI
>> devices.
>>
>>> Motherboard is an Asus M2N32-WS. DSDT attached.
>>
>> From a quick glance it should work. If it doesn't enable
>> CONFIG_HWMON_DEBUG_CHIP.
>
> Well, I just tried it here and it works here too, on 3 different
> asus motherboards.  But asus_atk0110 is far less useful than the
> it87 variant.  Yes atk0110 shows correct labels for various sensors,
> but for one there's no way to control fan speeds using it, at least
> not currently, -- something which is done by it87 easily.
>
> Well, one more difference is that atk0110 also shows vcore, which
> at least looks real instead of fake 1.55V as shown by it87.
>
> But anyway, loading the driver just to get correct labels and
> vcore seems a bit overkill...

The main reason for using atk0110 is correctness: the resources are
claimed by ACPI, it might not be safe to touch them (for the same
reason two drivers are not allowed to map e.g. the same PCI BAR).
On newer boards the risk of collision is pretty high, since the hwmon
chip is used by an EC that works in background... on other boards the
risk is much lower since the hwmon chip doesn't seem to be probed
actively.
Anyway, as user you can override this decision with
"acpi_enforce_resources=lax", but _I_ wouldn't recommend it.

Luca
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/