From: Laszlo Ersek [mailto:[email protected]
Sent: Monday, September 26, 2016 3:11 PM
To: Moore, Robert
Cc: devel(a)acpica.org; Michael Tsirkin; Box, David E
Subject: Re: Turning iasl's "-oe" option into a documented, regular
On 09/26/16 23:34, Moore, Robert wrote:
>> -----Original Message-----
>> From: Laszlo Ersek [mailto:[email protected]
>> Sent: Monday, September 26, 2016 2:23 PM
>> To: Moore, Robert <robert.moore(a)intel.com>
>> Cc: devel(a)acpica.org; Michael Tsirkin <mtsirkin(a)redhat.com>
>> Subject: Turning iasl's "-oe" option into a documented, regular
>> the addition of the External() opcode (i.e., 0x15), from ACPI v6.0,
>> causes iASL to generate AML that breaks old versions of ACPICA's AML
>> interpreter that are shipped with old (but still variably supported)
>> Linux distributions, for example, RHEL-5.11. The breakage happens
>> despite the If(0) conditional that surrounds the opcode and is
>> supposed to hide it.
Please explain further. The design of the External opcode was specifically chosen as to
not break existing interpreters.
>> I found ACPICA commit
>> commit 69620304a02d5a14d1fd0040aeb2e32bd1d7023e
>> Author: Robert Moore <Robert.Moore(a)intel.com>
>> Date: Fri Feb 26 08:55:38 2016 -0800
>> iASL: Add debug option for External opcode
>> option to temporarily disable the opcode for debug purposes
>> which adds the "-oe" option, but -- on purpose -- does not document
>> Is there any intention to turn this option into a regular, long-term,
>> documented option?
> [Moore, Robert]
> No, this is a temporary flag for debugging any issues. But the external
opcode support is stable, and we plan to remove the option.
Thank you for the quick response!
>> For reference, please see