On Wed, Jun 24, 2015 at 12:33 AM, Philip Guenther wrote:
> On Tue, Jun 23, 2015 at 1:18 PM, Mike Larkin wrote:
>> On Tue, Jun 23, 2015 at 09:49:26AM +0300, Alexey Dobriyan wrote:
>>> Clearing 32-bit register clears whole register, save REX prefix.
>>
>> I see nothing documented in the Intel SDM t
On Tue, Jun 23, 2015 at 02:33:51PM -0700, Philip Guenther wrote:
> On Tue, Jun 23, 2015 at 1:18 PM, Mike Larkin wrote:
> > On Tue, Jun 23, 2015 at 09:49:26AM +0300, Alexey Dobriyan wrote:
> >> Clearing 32-bit register clears whole register, save REX prefix.
> >
> > I see nothing documented in the
On Tue, Jun 23, 2015 at 1:18 PM, Mike Larkin wrote:
> On Tue, Jun 23, 2015 at 09:49:26AM +0300, Alexey Dobriyan wrote:
>> Clearing 32-bit register clears whole register, save REX prefix.
>
> I see nothing documented in the Intel SDM that says this. Can you cite
> a reference to support this claim?
On Tue, Jun 23, 2015 at 09:49:26AM +0300, Alexey Dobriyan wrote:
> Clearing 32-bit register clears whole register, save REX prefix.
I see nothing documented in the Intel SDM that says this. Can you cite
a reference to support this claim?
An aside - even if this were true, I still prefer xorq and
Clearing 32-bit register clears whole register, save REX prefix.
Index: arch/amd64/amd64/acpi_wakecode.S
===
RCS file: /cvs/src/sys/arch/amd64/amd64/acpi_wakecode.S,v
retrieving revision 1.35
diff -u -p -u -r1.35 acpi_wakecode.S
--- a