Re: [Xen-devel] [PATCH v2 03/23] x86: zero BSS using stosl instead of stosb

2015-07-22 Thread Jan Beulich
>>> On 22.07.15 at 13:22, wrote: > On 22/07/15 11:04, Jan Beulich wrote: > On 22.07.15 at 10:42, wrote: >>> In the case of having aligned source and destination on a 16-byte >>> boundary (which we can trivially arrange), then ERMSB (to give it its >>> Intel name) and rep stosl differ only in

Re: [Xen-devel] [PATCH v2 03/23] x86: zero BSS using stosl instead of stosb

2015-07-22 Thread Andrew Cooper
On 22/07/15 11:04, Jan Beulich wrote: On 22.07.15 at 10:42, wrote: >> In the case of having aligned source and destination on a 16-byte >> boundary (which we can trivially arrange), then ERMSB (to give it its >> Intel name) and rep stosl differ only in the setup cost; they still >> scale at t

Re: [Xen-devel] [PATCH v2 03/23] x86: zero BSS using stosl instead of stosb

2015-07-22 Thread Jan Beulich
>>> On 22.07.15 at 10:42, wrote: > In the case of having aligned source and destination on a 16-byte > boundary (which we can trivially arrange), then ERMSB (to give it its > Intel name) and rep stosl differ only in the setup cost; they still > scale at the same rate for changes in length. > > Th

Re: [Xen-devel] [PATCH v2 03/23] x86: zero BSS using stosl instead of stosb

2015-07-22 Thread Andrew Cooper
On 22/07/2015 06:18, Jan Beulich wrote: Daniel Kiper 07/21/15 8:23 PM >>> >> On Tue, Jul 21, 2015 at 03:37:48AM -0600, Jan Beulich wrote: >> On 20.07.15 at 16:28, wrote: >>> ... because of ??? Nowadays - with X86_FEATURE_ERMS - rep stosb >>> is expected to be faster than rep stosl. >> OK

Re: [Xen-devel] [PATCH v2 03/23] x86: zero BSS using stosl instead of stosb

2015-07-21 Thread Jan Beulich
>>> Daniel Kiper 07/21/15 8:23 PM >>> >On Tue, Jul 21, 2015 at 03:37:48AM -0600, Jan Beulich wrote: >> >>> On 20.07.15 at 16:28, wrote: >> >> ... because of ??? Nowadays - with X86_FEATURE_ERMS - rep stosb >> is expected to be faster than rep stosl. > >OK, I did not know about that. However, as I

Re: [Xen-devel] [PATCH v2 03/23] x86: zero BSS using stosl instead of stosb

2015-07-21 Thread Daniel Kiper
On Tue, Jul 21, 2015 at 03:37:48AM -0600, Jan Beulich wrote: > >>> On 20.07.15 at 16:28, wrote: > > ... because of ??? Nowadays - with X86_FEATURE_ERMS - rep stosb > is expected to be faster than rep stosl. OK, I did not know about that. However, as I know this feature was introduced in 2012 with

Re: [Xen-devel] [PATCH v2 03/23] x86: zero BSS using stosl instead of stosb

2015-07-21 Thread Jan Beulich
>>> On 20.07.15 at 16:28, wrote: ... because of ??? Nowadays - with X86_FEATURE_ERMS - rep stosb is expected to be faster than rep stosl. Jan ___ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel

[Xen-devel] [PATCH v2 03/23] x86: zero BSS using stosl instead of stosb

2015-07-20 Thread Daniel Kiper
Additionally, align relevant comment to coding style. Suggested-by: Andrew Cooper Signed-off-by: Daniel Kiper --- xen/arch/x86/boot/head.S |5 +++-- xen/arch/x86/xen.lds.S |2 ++ 2 files changed, 5 insertions(+), 2 deletions(-) diff --git a/xen/arch/x86/boot/head.S b/xen/arch/x86/boo