Re: [diff] trivia: change xorl => xorq

2015-06-24 Thread Alexey Dobriyan
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

Re: [diff] trivia: change xorl => xorq

2015-06-23 Thread Mike Larkin
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

Re: [diff] trivia: change xorl => xorq

2015-06-23 Thread Philip Guenther
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?

Re: [diff] trivia: change xorl => xorq

2015-06-23 Thread Mike Larkin
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

[diff] trivia: change xorl => xorq

2015-06-23 Thread Alexey Dobriyan
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