ACPI requesting I/O ports?
Hi Len,
As a follow up of this dicussion:
http://lists.lm-sensors.org/pipermail/lm-sensors/2007-March/019015.html
I'd like to know if it would be possible to have the acpi subsystem somehow declare or even request the I/O ports it will be accessing when running AML code? There is a need for this, otherwise other drivers might access the same resources as the AML code does, with possibly bad consequences.
Another approach proposed by Pavel Machek would be a general mutex to protect the AML code from the rest of the kernel, this would probably work too but the granularity (and thus the latency) would be better with a per-I/O-region approach.
Thanks,
Signature · 1 line
Jean Delvare
Port addresses can be dynamically generated by the AML code and thus, there is no way that the ACPI subsystem can statically predict any addresses that will be accessed by the AML.
Quoted 8 lines
-----Original Message-----
From: [email protected] [mailto:linux-acpi- [email protected]] On Behalf Of Jean Delvare
Sent: Friday, March 02, 2007 1:15 PM
To: Brown, Len
Cc: [email protected]; Pavel Machek
Subject: ACPI requesting I/O ports?
Quoted 4 lines
Hi Len,
As a follow up of this dicussion:
http://lists.lm-sensors.org/pipermail/lm-sensors/2007-March/019015.html
Quoted 3 lines
I'd like to know if it would be possible to have the acpi subsystem somehow declare or even request the I/O ports it will be accessing
when
Quoted 2 lines
running AML code? There is a need for this, otherwise other drivers might access the same resources as the AML code does, with possibly
bad
Quoted 7 lines
consequences.
Another approach proposed by Pavel Machek would be a general mutex to protect the AML code from the rest of the kernel, this would probably work too but the granularity (and thus the latency) would be better with a per-I/O-region approach.
Quoted 5 lines
Thanks,
--
Jean Delvare
-
To unsubscribe from this list: send the line "unsubscribe linux-acpi"
in
Quoted 2 lines
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Hi Robert,
Quoted 3 lines · Robert
Port addresses can be dynamically generated by the AML code and thus, there is no way that the ACPI subsystem can statically predict any addresses that will be accessed by the AML.
Can you please explain in greater details how "dynamical" it is? Any real world example you could present?
Also, you say that port addresses _can_ be dynamically generated, which suggests that some are not, so the ACPI subsystem could at least properly request these ones?
Thanks,
Signature · 1 line
Jean Delvare